Workstream 1 – economic evaluation

Decision problem

Aim

The aim of the independent economic evaluation is to assess the cost-effectiveness of different strategies for selecting hospitalised patients for pharmacological thromboprophylaxis including prophylaxis for all, prophylaxis for none and prophylaxis based on a VTE RAM. Sections of this chapter have been reproduced from Pandor et al. under licence CC BY 4.0.74

Population

The target population is hospital inpatients, including medical, surgical and trauma patients but excluding critical care patients, children and women admitted to hospital for pregnancy-related reasons. In order to allow for differences in patient characteristics and risks within this very broad population, the economic analysis considered two broad groups as follows:

• acutely ill medical inpatients
• surgical inpatients including both elective and emergency surgery.

The analysis focused on those populations where NG89 (NICE Guideline 89: Venous thromboembolism in over 16s: reducing the risk of hospital-acquired DVT or PE)8 recommends the use of thromboprophylaxis based on the balance of VTE and bleeding risks. The analysis is not expected to be applicable to very specialist patient groups, such as neurosurgical patients, where the decision whether to use prophylaxis is based on a very individualised approach and the use of a RAM designed for general hospitalised patients is unlikely to be relevant.

Patients identified to be at increased risk of bleeding, such as those identified on the Department of Health risk assessment tool, are assumed not to receive pharmacological thromboprophylaxis under any strategy and are excluded from the model. Therefore, when we refer to a strategy of ‘prophylaxis for all’, this does not mean offering pharmacological thromboprophylaxis to those identified as being at high risk of bleeding.

We considered whether to model elective and emergency surgical patients separately. However, one of the more established risk-scoring tools for surgical patients, Caprini, has been validated in a mixed population of elective and emergency surgical patients and includes questions that would differentiate emergency surgical patients, such as those who have had fractures, into higher-risk groups. It therefore seemed reasonable to evaluate RAMs for use across the broad category of surgical inpatients and to treat the reason for surgery as a risk factor rather than modelling separate decision-making processes in elective and emergency surgical inpatients.

Trauma patients are not modelled as an explicit subgroup as trauma patients are highly complex, often have critical site bleeding injuries and are not managed within a specific trauma speciality in the UK. Managing speciality is often determined by their main complaint, which means they usually fall under a neurosurgical, spinal, general or orthopaedic speciality and, therefore, fall within the emergency surgical population. In addition, many of the RAMs which have been developed specifically for trauma patients, have been developed and validated in countries where trauma patients are treated within a critical care environment; this cohort is excluded from project scope.

Patients having elective hip or knee replacements are examined as a specific subset within the surgical inpatient cohort because of procedure-specific national recommendations (NG89); these patients are recommended to have a longer duration of prophylaxis than is typical in other surgical populations.

Strategies for prophylaxis

The prophylaxis interventions examined were chosen based on the recommendations in NG89 as these were selected based on cost-effectiveness modelling and, therefore, should represent the most cost-effective strategies. For most types of surgery, NG89 recommends that low-molecular-weight-heparin (LMWH) is considered if the person’s VTE risk outweighs their risk of bleeding. Although the use of mechanical prophylaxis in the form of antiembolism stockings (AES) is generally recommended for surgical patients within NG89, a recent study has shown that LMWH alone is non-inferior to LMWH with AES in elective surgical patients who were assessed as being at moderate or high risk of VTE.75 Given that it is the use of LMWH that is being driven by risk assessment and not the use of AES, the comparison of interest for the modelling is LMWH versus no LMWH. The conclusions from the modelling should apply equally to patients having mechanical prophylaxis or no mechanical prophylaxis. This is because the model estimates whether it is cost-effective to use a RAM to decide whether to add LMWH to the treatment of patients who would not otherwise receive it and that is not dependent on any underlying care being given, provided that the underlying care is not altered by the addition of LMWH. Further discussion of mechanical thromboprophylaxis was, therefore, considered to be outside the scope of the modelling.

For acutely ill medical patients, LMWH is recommended in NG89 but mechanical prophylaxis is not recommended in acutely ill medical patients. Therefore, for acutely ill medical patients we compare LMWH to no prophylaxis.

When LMWH is used, we assume that it is given for the duration of admission. Although NG89 recommends that LMWH is given for a minimum of 7 days, information from the National VTE Exemplar Centres Network suggests that a pragmatic approach has been taken to the implementation of NG89 with LMWH generally not being prescribed for post-discharge use.76 Therefore, we assume in the base-case analysis that LMWH is given for the duration of admission, but we also explore a scenario analysis in which it is given for 7 days even if this means post-discharge administration.

In higher-risk populations, such as elective hip and knee replacement, an extended form of pharmacological prophylaxis is recommended in NG89. Therefore, we have modelled these groups as specific subsets of the surgical population in separate analyses, described in Model inputs for the elective hip replacement and elective knee replacement scenarios, to determine if the threshold for cost-effective intervention is different in these groups. In addition, NG89 recommends that clinicians should ‘consider extending pharmacological VTE prophylaxis to 28 days postoperatively for people who have had major cancer surgery in the abdomen’. Therefore, a scenario analysis is conducted to determine if the threshold for cost-effective intervention is different in general surgical patients having 28 days of LMWH.

In summary, the main comparisons being modelled are:

• acutely ill medical inpatients – LMWH for duration of admission versus no prophylaxis
• surgical inpatients (excluding elective hip and knee replacement) – LMWH for duration of admission (without or without AES) versus no pharmacological prophylaxis (with or without AES).

Although the recommendations in NG89 were informed by modelling to determine the most cost-effective thromboprophylaxis strategy, the analysis did not explore the use of RAMs to determine which patients should be offered prophylaxis. The RAMs have been selected based on the review in the previous chapter. For medical patients, the primary analysis compares a group of several RAMs that have been externally validated within a single cohort as this minimises the risk of introducing bias from comparing tools that have been studied in populations with different characteristics. A secondary analysis then compares the remaining tools that have been externally validated in different medical cohorts. The number of RAMs for surgical patients was more limited. Therefore, for surgical patients we have included in the primary analysis all tools externally validated in either a surgical or mixed medical and surgical population. In all analyses, the comparator strategies of LMWH for all and LMWH for none are included. The tools evaluated are summarised in Table 4.

Modelling methods

Context

The model estimates lifetime costs and QALYs for the different thromboprophylaxis strategies and the comparator of no thromboprophylaxis under a NHS and Personal Social Services perspective. Future costs and benefits are both discounted to their net present value at a rate of 3.5% per annum in accordance with the NICE guide to the methods of technology appraisal.77 Costs are reported in Great British pounds based on 2020 prices. To achieve this, historical prices used as model inputs were uplifted using the hospital and community health services pay and prices index up to 2016 and the NHS cost inflation index thereafter.78

Conceptual model for medical inpatients

The model consists of a decision tree (see Figure 5) to capture short-term outcomes followed by a lifetime state transition (Markov) model (see Figure 6) to capture the impact of outcomes that result in death or ongoing morbidity. The model estimates outcomes for a cohort of identical patients with average characteristics. In reality, the application of RAMS may lead to treated and untreated patients having different characteristics. Based on clinical advice, age was the patient characteristic considered most likely to result in variation in outcomes within the cohort being risk assessed. We conducted a scenario analysis to explore whether incorporating patient heterogeneity into the model would affect the conclusions by varying the age of the cohort. From this we concluded that the cohort-level modelling approach was adequate and would not introduce significant bias as the optimal strategy was fairly robust to changes in the age.

FIGURE 5 Short-term (6-month) decision tree model structure. (Figure not reproduced here; see the original article.)

FIGURE 6 Long-term Markov model. (Figure not reproduced here; see the original article.)

The decision tree is used to estimate for each strategy: the number of patients receiving thromboprophylaxis, the impact of thromboprophylaxis on VTE outcomes (PEs and DVTs) and the incidence of major bleeds during either thromboprophylaxis or VTE treatment with anticoagulants. Major bleeds were considered to be those meeting the International Society for Thrombosis and Haemostasis (ISTH) definitions for major bleeding for surgical and non-surgical (medical) patients.79,80 Major bleeds were divided into fatal bleeds, non-fatal intracranial haemorrhages (ICHs) and other major bleeds (referred to as non-fatal non-ICH major bleeds). Gastrointestinal (GI) bleeds were assumed to be a reasonable proxy for non-fatal non-ICH major bleeds and these were assumed to have no impact on costs or quality of life after 1 month. PEs were divided into fatal and non-fatal events. DVTs were divided first into symptomatic and asymptomatic DVTs and then into proximal and distal DVTs. Symptomatic DVTs and non-fatal PEs are assumed to result in 3 months of anticoagulant treatment. Patients having major bleeds during either prophylaxis or VTE treatment are assumed to stop their anticoagulant medication at the time of the bleed. In our previous analysis of thromboprophylaxis strategies in patients having lower limb immobilisation following injury, we found that the prevention of post-thrombotic syndrome (PTS) following asymptomatic DVT was an important driver of both cost-effectiveness and decision uncertainty due to asymptomatic DVTs being more common than symptomatic DVTs but their long-term consequences being more uncertain.74 So while asymptomatic DVTs are assumed to remain undetected and untreated, it is important to capture these DVTs in the decision tree phase of the model in order to capture any ongoing morbidity due to PTS in the long-term state-transition model. In the decision tree, any VTE event that is related to the hospital admission is assumed to occur within 90 days of hospital admission. Therefore, a 6-month time frame was considered sufficient to capture the period of VTE risk associated with the hospital admission and the period of treatment following VTE (3 months). Diagnosis of PTS and chronic thromboembolic pulmonary hypertension (CTEPH) is assumed not to occur until the end of the decision-tree phase of the model, as it is difficult to distinguish these chronic complications from acute symptoms during the first 3 months after VTE. The decision tree allows for the possibility that patients not having anticoagulation may have a major bleed during their hospitalisation. The likelihood of VTE and the likelihood of bleeding during treatment for VTE is assumed to be independent of patient characteristics (e.g. age and sex) and independent of whether the patient had major bleeding during hospitalisation. Clinically relevant non-major bleeding (CRNMB) is excluded in the base-case analysis as it was considered that CRNMB occurring during hospitalisation would have a minimal impact on costs and quality of life in patients already hospitalised. Heparin-induced thrombocytopenia (HIT) was not included in the model because the most important consequence of HIT is the increased risk of VTE, but any increase in VTE related to HIT in the LMWH arms of the clinical trials would be included within the efficacy estimates for LMWH versus placebo and would, therefore, already be accounted for in the model.

The key model assumptions for the decision tree phase are as follows:

• Bleeding events following hospitalisation are possible in both those having thromboprophylaxis and those having no thromboprophylaxis.
• Bleeds associated with thromboprophylaxis are assumed to occur during prophylaxis and, therefore, before hospital discharge.
• VTE associated with hospitalisation is assumed to occur within 90 days of admission.
• Patients who have major bleeding will stop thromboprophylaxis immediately, but the treatment effect of thromboprophylaxis is assumed to be the same as for those who completed treatment, as patients who bleed are assumed to be adequately anticoagulated.
• The risk of VTE is the same whether or not thromboprophylaxis caused bleeding.
• All patients with symptomatic DVT receive accurate diagnosis and initiate treatment with anticoagulants [3 months of either direct oral anticoagulants (DOACs) or phased anticoagulation].
• Asymptomatic DVTs are not detected and are not treated.
• All PEs are symptomatic and lead to detection and treatment (3 months of either DOACs or phased anticoagulation) in all cases.
• Patients treated for symptomatic DVT and PE have a bleed risk associated with treatment which is assumed to occur during the 3-month treatment period.
• Chronic complications of VTE (CTEPH following PE and PTS following DVT) are assumed to be diagnosed at least 3 months after VTE and therefore occur after any bleeds associated with VTE treatment.
• Deaths due to PE occur before any bleeding associated with the treatment of PE.
• Risk of bleeding during treatment of VTE is independent of whether the patient bled during prophylaxis.
• Risk of VTE, risk of bleeding and risk of PTS/CTEPH are based on average patient characteristics (e.g. age and sex) for the cohort being risk assessed.

A state-transition model (see Figure 6) was then used to extrapolate lifetime outcomes including overall survival and ongoing morbidity related to either bleeds or VTE. The health states included within the state-transition model capture the risk of PTS following VTE and the risk of CTEPH following PE. The risk of PTS is dependent on whether the DVT is symptomatic and treated or asymptomatic and untreated, and also whether the DVT is proximal or distal. All patients with PTS are combined in a single health state as costs, utilities and survival are not expected to be affected by whether PTS occurred following proximal or distal DVT. The PTS health state is not split into different severity levels as the utility estimates are based on the average utility across severity levels and the costs are not expected to differ by severity. The CTEPH health state is divided according to whether patients receive medical or surgical management to allow for differential costs and survival between these groups. There is also a post-ICH state to capture ongoing morbidity following ICHs. Further adverse outcomes (PTS, CTEPH) are not modelled following ICH, as lifetime costs and QALYs are assumed to be predominantly determined by morbidity related to ICH. Recurrent VTEs are not modelled as it is unclear whether any recurrent VTE would be related to the VTE provoked by hospital admission or whether it is related to underlying risk factors and would therefore have occurred whether or not prophylaxis was initiated during the hospital admission. The state-transition model has one 6-month cycle to extrapolate the outcomes of the decision tree up to 1 year, followed by annual cycles thereafter. All-cause mortality during the first year is applied at exactly 6 months. Thereafter, the health state occupancy is half-cycle corrected such that all transitions between states, including mortality, are assumed to occur mid-cycle. The key model assumptions during the state-transition phase are as follows:

• All symptomatic DVTs are associated with a risk of PTS but the rate is allowed to differ depending on whether the DVT is distal or proximal and whether it is symptomatic and treated or asymptomatic and untreated.
• There is no risk of PTS following PE and CTEPH is possible only after PE.
• Further outcomes (i.e. VTE, CTEPH and PTS) are not modelled for those who experience ICH as lifetime cost and QALYs will be determined predominantly by disability related to the ICH.
• All-cause mortality is applied to all transition states except CTEPH and post-ICH which have state-specific mortality rates.
• Recurrent VTE is not modelled.

Conceptual model for surgical patients

We considered whether it was necessary to add additional wound-related outcomes to the model for surgical patients, such as surgical site bleeding or surgical site infections associated with wound haematoma. The clinical experts advised that the vast majority of major bleeds in surgical patients are distant from the surgical site. In addition, any complications related to surgical site bleeding that require patients to return to theatre or that result in prolonged hospitalisation would fall under the ISTH definition of major bleeding in surgical patients.79 Therefore, these would already be captured within the decision-tree model as non-fatal non-ICH major bleeds. For these reasons, the conceptual model for surgical patients was kept identical to that for medical inpatients, with the caveat that the definition of major bleeding is different for surgical patients. Extended duration thromboprophylaxis in clinical subgroups (such as surgical patients with abdominal cancer) was evaluated using the same conceptual model with a separate scenario analysis (as per Strategies for prophylaxis).

Data sources

The input parameters used in a previous analysis of thromboprophylaxis during lower limb immobilisation were examined to identify any that were less relevant to patients receiving prophylaxis during hospitalisation. The data related to the population characteristics (age, sex and life expectancy), the incidence of VTE, the incidence of bleeding and the costs of prophylaxis were updated to use data specific to our target population. Other data such as the incidence of chronic sequelae (PTS and CTEPH) following VTE, and the costs and utility values for patients experiencing adverse outcomes were generally based on the same sources with costs updated to reflect changes in prices. One exception is the utility data for DVT and PE which have been updated because a more detailed analysis of the study used previously is now available.

Clinical input parameters are described below in Population characteristics to Mortality risks for adverse outcomes. Those that are specific to the medical and surgical inpatient populations are summarised in Table 5, with full details of all clinical input parameters provided in Appendix 2, Table 23. Cost inputs are described below in Cost of prophylaxis to Costs of managing PTS and CTEPH and summarised in Appendix 2, Table 24 with a detailed breakdown of drug costs for treating VTE and resource use for diagnosing and treating VTE provided in Appendix 2, Tables 25 and 26. Utility inputs are described below in Utility values and summarised in Appendix 2, Table 27 for the decision tree phase of the model and Appendix 2, Table 28 for the state-transition phase of the model. Probabilistic distributions for cost and utility inputs are summarised in Appendix 2, Table 29. The assumed timings of utility inputs are described in Timing and duration of utility decrements applied in the decision tree. Sensitivity and specificity data for the alternative RAMs are described in Sensitivity and specificity of RAMs.

Population characteristics

The cohort of patients reported in the papers by Greene et al. and Grant et al. was selected as being representative of the medical inpatient population.38,39 This cohort was selected as it is the cohort used to estimate the sensitivity and specificity for five RAMs in the primary analysis for medical inpatients and the inclusion criteria were not considered to be overly restrictive compared with the target population. The starting age in the model was, therefore, set to be 65.8 years of age with 44.5% of the starting population being male.

The pooled mean age of surgical patients in the UK in 2015 was reported by Fowler et al. to be 54.2 years based on analysis of hospital episode statistics (HES) data.81 This sample included both day-case procedures (45.6%) and regular inpatient admissions and did not distinguish between elective and emergency surgery. Of these 45.8% were surgical admissions in men. A cohort study examining emergency surgical admissions across six UK acute surgical units reported a mean age of 54 with 56% being female.82 Based on these two sources we have assumed that the surgical cohort being modelled has an average age of 54 years of age and that 46% are male.

All-cause mortality

Life expectancy is known to vary between men and women of the same age, and therefore any QALY gains from deaths prevented would be dependent on sex. Rather than modelling male and female patients separately to account for their different mortality risks, we used age and sex-specific general population mortality data to calculate the average risk of mortality for the cohort as a whole based on their age at baseline and the proportion of males and females at baseline. The weighting of the mortality risk across males and females was allowed to vary over the course of the model to allow for the fact that lower mortality risks in females lead to a slight increase in the proportion alive who are female over time.

Having calculated the all-cause mortality for an age- and sex-matched cohort with general population mortality risks, it was then necessary to consider the increased risk of mortality for hospitalised patients relative to the general population. A study by Moore et al. reported that medical inpatients have a ninefold risk of mortality in the year following admission when compared to an age- and sex-matched general population cohort.83 We have therefore applied relative risk (RR) of 9.4 to the estimate of general population all-cause mortality in the first year of the model. This gives a 10% absolute risk of death in the first year of the model, which is substantially lower than the absolute mortality rate observed by Moore et al., which was 22.4% at 1 year. However, it is unclear whether the absolute risk of mortality reported by Moore et al. is representative of current practice across the UK. The 30-day risk reported by Moore et al. is 6.5%, which is higher than the 30-day mortality rate for non-elective inpatients reported in the Summary Hospital-level Mortality Indicator (Standard Hospital Mortality Indicator) tables for England, which was 3.4% in 2018/19.84 We have conducted a sensitivity analysis in which an absolute mortality rate of 22.4% is applied in the first year of the model for all-cause mortality to see whether uncertainty around all-cause mortality is important in determining the optimal thromboprophylaxis strategy.

In the surgical population, the RR from Clark et al. for all-cause mortality in the year after hospitalisation for medical versus surgical patients (RR = 1.9, 95% CI 1.7 to 2.0) is applied to estimate the all-cause mortality in surgical patients.85 This gives a RR for surgical patients versus the general population of 4.9 in the first year after hospitalisation. Clark et al. did not distinguish between emergency and elective surgery. A comparison of crude 30-day mortality rates for elective and emergency admissions, suggests that mortality is lower for elective admissions (1% vs. 3.4%)84 but as the emergency admissions are not reported separately for medical and surgical patients it is unclear how much of this is due to the inclusion of acutely ill medical patients in the emergency admission cohort. To explore whether a lower risk of mortality in elective surgical patients could lead to different recommendations for thromboprophylaxis, we have conducted a sensitivity analysis in which we apply general population mortality rates to surgical patients.

Moore et al. also present information on the mortality risk over the course of the first year and this shows that the rate of mortality is high initially and then decreases over time, although they do not report whether it reaches general population levels of mortality by the end of the first year. To allow for the fact that the risk of all-cause mortality decreases with time since discharge, we have not included any increased risk of mortality in year 2 and beyond for hospitalised medical and surgical patients relative to the general population.

Risk of VTE in medical inpatients

The risk of VTE in medical inpatients not receiving pharmacological prophylaxis has been taken from a prospective observational study by Barbar et al., which was the derivation cohort for the Padua RAM.15 The paper reports the rate of VTEs in high- and low-risk patients according to whether or not they received prophylaxis. Therefore, it is possible to estimate the risk of VTE in those not receiving prophylaxis across both low and high-risk patients. However, a weakness of this study design is that prophylaxis is more likely to be offered to higher-risk patients. Therefore, this study does not give an estimate of the rate of VTEs in an unselected population who do not receive prophylaxis. However, only 39.7% of the higher-risk group (defined as those with a Padua score ≥ 4) received prophylaxis. Higher-risk patients not receiving prophylaxis were younger, less likely to have reduced mobility, were less likely to have heart and/or respiratory failure and were less likely to have acute infection or rheumatological disorder. Despite this the risk of VTE in higher-risk patients not receiving prophylaxis was 11.8%, which is considerably higher than the risk of 2.2% in higher-risk patients receiving prophylaxis. We cannot know what the risk would have been in the latter group had prophylaxis not been administered. However, the majority of the patients in this cohort did not receive prophylaxis, even in the higher-risk group, and this paper allows us to estimate the risks of VTE in those not receiving prophylaxis. The risks of VTE in those not receiving prophylaxis within the cohort report by Barbar et al. was 1.38% for PE, and 2.02% for symptomatic DVT, giving a total risk of 3.40% for symptomatic VTE.

Using data from this prospective cohort was considered preferable to using data from the placebo arms of randomised controlled trials (RCTs) for two reasons. Firstly, many of the RCTs comparing prophylaxis against placebo are old and may not be representative of the risks of VTE within current clinical practice. Secondly, many RCTs have restrictive inclusion and exclusion criteria which results in a selective population that may over or under-estimate the risk of VTE in real-world clinical practice. The rates of VTE from the subset of patients reported by Barbar et al. who did not receive prophylaxis were applied in the base case but scenario analyses exploring higher rates of VTE are also presented as it is unclear whether the risk of VTE would be higher in an unselected population not receiving prophylaxis.

Barbar et al. did not conduct any form of screening to identify asymptomatic DVT. The economic model which informed the 2009 NICE guidelines on VTE prevention in hospitalised patients (CG92) used the ratio of asymptomatic to symptomatic DVTs from those RCTs reporting both distal and proximal asymptomatic DVTs across all patient groups excluding total hip and total knee replacement.86 This gave a ratio of 40 symptomatic DVTs to 604 asymptomatic DVTs. From this we calculated that the incidence of asymptomatic DVTs in the cohort reported by Barber et al. would be 30.46%. Therefore, the overall rate of VTE including asymptomatic DVTs would be 33.85%. We also needed to determine the proportion of DVTs which are proximal as distal DVTs are assumed to have different consequences, such as a lower risk of PTS. In the model that informed CG92, the ratio of distal to proximal DVTs was based on all RCTs included in their systematic review which reported the incidence of both. Based on this, the proportion of all DVTs which are proximal was estimated to be 31% and this was assumed to be constant across both symptomatic and asymptomatic DVTs.

The RR of VTE was estimated from the RCTs included in the review for NG89, which reported relevant outcomes.8 For acutely ill medical inpatients there are 12 studies described in NG89. Of these only four compared LMWH (standard dose/standard duration) to placebo,87–90 but one of these only reported mortality outcomes89 (all-cause mortality and fatal-PE), leaving three studies reporting VTE outcomes. All three RCTs reported the incidence of VTE including both PE and DVT. Two of these studies88,90 included both symptomatic DVT and asymptomatic DVT in their VTE incidence, although one included only asymptomatic proximal DVT.88 The third study87 did not assess patients for asymptomatic DVT and, therefore, only reported symptomatic DVT or PE. We conducted a random-effects meta-analysis of VTE outcomes from these three studies to obtain an estimate of the RR (0.49, 95% CI 0.37 to 0.67) and applied this to all VTE outcomes within the model. Details of the numbers analysed are included in Appendix 2, Figure 27.

Risk of VTE in surgical inpatients

The risk of VTE in surgical inpatients was taken from the risk reported by Pannucci et al. in the derivation cohort for the Pannucci RAM.54 This cohort was chosen as PE rates (0.62%) were reported separately in addition to the overall incidence of VTE at 90 days (1.40%). The overall incidence of VTE in the validation cohort, reported by Pannucci et al., and in the validation cohort for the Caprini RAM, reported by Bahl et al. were similar to that reported in the derivation cohort by Pannucci et al. (1.40% at 90 days and 1.44% at 30 days, respectively), but details regarding the split between PE and DVTs were not provided for these validation cohorts.26,54 The data from the Pannucci et al. cohort were in an exclusively non-emergent surgical population, whereas Bahl et al. included both elective and emergency surgical patients, but it seems reasonable to use the data from Pannucci et al. for all surgical patients, given that the overall rate of VTE is similar in Bahl et al. In both the derivation and validation cohorts reported by Pannucci et al., a similar proportion of the cohort received no pharmacological prophylaxis (34% in both cases). However, the risk of VTE is not reported separately for those not receiving prophylaxis by Pannucci et al., meaning that the risk of VTE in a cohort receiving no prophylaxis may have been underestimated. Bahl et al. did attempt to adjust for adherence to the local prophylaxis protocol, but they only reported what the VTE rate would have been for a cohort with full adherence, rather than the rate of VTE expected without any thromboprophylaxis. Therefore, the incidence from Pannucci et al. was applied in the model, albeit with the limitation that the use of pharmacological prophylaxis in around two-thirds of cases in this cohort may have underestimated the risk in a cohort without any pharmacological prophylaxis. Given this limitation, sensitivity analyses were conducted to explore whether the most cost-effectiveness strategy for pharmacological prophylaxis would be different if the VTE level risk was higher or lower than observed by Pannucci et al. The risk of asymptomatic DVT was estimated using the same assumption as applied in medical inpatients (i.e. that 6.2% of all DVTs are symptomatic), giving a risk of asymptomatic DVT of 12.6%. The same ratio of proximal to distal DVTs as applied in medical inpatients was also applied to elective surgical inpatients (i.e. 31% of all DVTs are proximal).

The RR of VTE in surgical inpatients for LMWH in combination with AES compared with AES alone was not readily available from the review reported in NG89. Separate network meta-analyses (NMA) are reported for elective hip replacement, elective knee replacement and abdominal surgery in NG89, with the latter being considered most generalisable to our target population (as discussed in Population). However, the NMA conducted for NG89 separated studies according to the timing and duration of LMWH, and separated studies according to the exact form of mechanical prophylaxis used meaning that the efficacy estimates for individual prophylaxis strategies are based on very few studies. As an alternative to using the data from NG89, we used data from a systematic review and economic analysis by Wade et al. which aimed to determine the relative effectiveness of knee versus thigh length AES when used in combination with pharmacological prophylaxis.91 The base-case NMA conducted by Wade et al. made the simplifying assumption that the treatment effect for heparin with AES versus AES alone would be similar to the treatment effect of heparin versus no prophylaxis. This assumption is in part supported by the recent GAPS study, not included in the NMA but which showed no clinical benefit from adjunctive graduated compression stockings in addition to LMWH in moderate risk surgical patients.75,92 Wade et al. also combined studies using LMWH, UFH and low-dose heparin within the network into a single heparin comparator. Both of these simplifying assumptions had the effect of allowing the estimate of treatment effect to draw on more studies than the approach used in NG89, which tended to split studies according to the exact prophylaxis regimen rather than lump them together. In addition, the analysis by Wade et al. did not distinguish between different types of surgical patients in their base case. Although scenario analyses were conducted excluding orthopaedic surgery, the economic analysis assumed that the efficacy estimates were common across all surgical patients. Wade et al. report that the median OR for heparin versus no heparin was 0.26 [95% credible interval (CrI) 0.09 to 0.87] in the whole surgical population for the outcome of DVT. In the economic model reported by Wade et al., the efficacy estimates for DVT were applied to all VTE outcomes as there was no evidence of differential effects by VTE event type (i.e. DVT and PE). Similarly, we have applied this efficacy estimate for all VTE outcomes in surgical patients.

Risk of major bleeding during prophylaxis in medical inpatients

When estimating the risk of major bleeding in the model, the absolute risk in those having LMWH is likely to be estimated more precisely from empirical data sources than the absolute risk in those not having prophylaxis as it will be based on more events. We, therefore, decided to estimate the absolute risk of major bleeding in those having LMWH in the model and then use a RR to estimate the risk in those having no prophylaxis.

The current NICE guideline for preventing VTE in hospitalised patients (NG89) provides an estimate of the risk of major bleeding for patients randomised to either LMWH or placebo based on 4051 patients enrolled in three RCTs.87,88,90 The RR of major bleeding for LMWH compared to placebo (1.53, 95% CI 0.8 to 2.92) was taken from the meta-analysis of three RCTs provided in N89 (Table 88 of NG89).

The three RCTs included in NG89 gave a risk of 1.02% for patients receiving LMWH. We considered whether it would be better to use data from a prospective cohort study such as the study by Barbar et al. which was used to estimate the absolute risk of VTE in those having prophylaxis.15 However, the number of patients having prophylaxis in the Barbar et al. cohort was only 238, making the estimate of risk less precise than that provided by the RCTs included in NG89. In addition, the concern that patients enrolled in RCTs would be a selective group was considered less important for bleeding outcomes as inclusion/exclusion criteria for RCTs would be more likely to focus on VTE risk than bleeding risk. Therefore, we decided to use the data from NG89 based on the three RCTs to estimate bleeding risks in medical inpatients in the model.

The three RCTs used to estimate the risk of major bleeding did not provide sufficient information to determine the breakdown of major bleeding events into fatal bleeds, non-fatal ICHs and other major bleeds. For this information we used data from the IMPROVE registry, which was a multinational, observational study designed to examine VTE prophylaxis patterns and clinical outcomes in patients hospitalised with an acute medical illness.93 Decousus et al. report bleeding outcomes from 10,866 patients of which 5231 (48.1%) received pharmacological prophylaxis including 4172 (38.4%) who received LMWH. The risk of major bleeding was 1.2% in the IMPROVE cohort overall. The type of bleeding is not reported according to whether or not prophylaxis was given or the type of prophylaxis used. However, the study does report data in sufficient detail to calculate the proportion of major bleeds that were fatal (14%) and the proportion of non-fatal major bleeds that were ICHs (10%). We have, therefore, assumed that these proportions are constant regardless of whether the bleeding occurred during LMWH or in patients not receiving LMWH.

Risk of major bleeding during prophylaxis for surgical inpatients

In NG89, the risk of bleeding for LMWH in surgical patients was estimated from five RCTs which compared LMWH either to placebo or to mechanical prophylaxis (which is not thought to increase the risk of bleeding) in patients having abdominal surgery.94–98 The absolute risk of major bleeding across the 297 patients who received LMWH across the five RCTs was 3.7% (11/297). Neither of the studies reporting validation cohorts for RAMs in surgical populations reported the risk of bleeding in the population receiving prophylaxis.26,54 Therefore, the data from NG89 for patients having abdominal surgery were used to inform the absolute risk in those receiving LMWH.

In NG89, the RR for major bleeding in patients having abdominal surgery obtained from the NMA was 2.98 (95% CI 0.88 to 14.80) for LMWH (postoperatively, standard duration, standard dose) versus placebo or mechanical prophylaxis. This RR was applied in the model for surgical patients.

The five RCTs used to estimate the absolute risk of major bleeding in surgical patients do not provide consistent information on the site of bleeding making it difficult to estimate the proportion of major bleeds that are ICHs. The data from the IMPROVE registry, which was conducted exclusively in medical inpatients, were not considered applicable to patients having surgery due to differences in the ISTH definitions of major bleeding for surgical and non-surgical patients.79,80

We identified a number of registries/cohort studies that reported the rate of major bleeding in surgical patients receiving thromboprophylaxis through ad hoc searches using Google Scholar. All of the identified registry/cohort studies recruited either elective hip replacement patients,99 elective knee replacement patients100 or both.101–103 As discussed in Population, we decided to model the use of thromboprophylaxis in patients having elective hip or knee replacement separately as the recommended duration of prophylaxis is greater in these patients than in most other surgical groups. However, we were unable to find any registry studies describing the site of major bleeding (i.e. ICH vs. non-ICH major bleeding) in non-orthopaedic populations having thromboprophylaxis. Our clinical experts advised that most major bleeds in surgical patients are distant from the surgical site and, therefore, information on the proportion of major bleeds that are fatal or ICHs is likely to be transferable from the orthopaedic population to the non-orthopaedic population.

From the registry/cohort studies identified, we chose to use the data from the XAMOS Xarelto^® in the prophylaxis of post-surgical venous thromboembolism after elective major orthopedic surgery of hip or knee (XAMOS) cohort study103 because it provided a detailed breakdown of the number of patients meeting various criteria for major bleeding, which allowed us to identify which bleeds would fall within the ISTH definition of major bleeding. We did this by taking the number meeting the EMA definition (122 for the adjusted safety population) and removing patients whose only reason for being categorised as having major bleeding was bleeding leading to treatment cessation (13 for the adjusted safety population) as this is not part of the ISTH criteria. The XAMOS study compared rivaroxaban to ‘standard care’, which included a variety of different thromboprophylaxis interventions, but a high proportion (81.7%) received LMWHs. Therefore, the major bleeding events in the standard-of-care arm were considered to be broadly representative of what would be expected for patients receiving LMWH. From the data reported from the XAMOX cohort, we estimated that 0.9% (1 in 109) of major bleeds meeting the ISTH definition are fatal and that 1.9% of non-fatal major bleeds are ICH (2 in 108) in patients having orthopaedic surgery. The proportion of non-fatal non-ICH major bleeds requiring reoperation is 13%. These proportions were applied in the model to the number of major bleeds for surgical patients to estimate the number of fatal bleeds and non-fatal ICHs.

Risk of major bleeding during treatment for VTE

The risk of bleeding during treatment for VTE has been taken from a paper by Kooiman et al. which reports the incidence of bleeding at 6 months in a cohort of patients having treatment for VTE [LMWH followed by vitamin K antagonist (VKA)] stratified by Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile international normalised ratio, Elderly, Drugs/alcohol concomitantly (HAS-BLED) score.104 This was used to calculate the incidence in those with a HAS-BLED score of 0 to 1 which was 0.9% (3/335). The rationale for using data from a population with a low HAS-BLED score is that inpatients would be unlikely to be considered for prophylaxis if they had significant risk factors for bleeding and, therefore, the population with a low HAS-BLED score is more representative of the population being modelled than an unselected population. The risk in the whole cohort starting treatment from the same study was 2%. For comparison, the risks in other unselected cohorts were 1.5% at 6 months in the Prevention of Thromboembolic Events – European Registry in Venous Thromboembolism (PREFER-VTE) registry and 2.24% at 3 months in the Computerized Registry of Patients with Venous Thromboembolism (RIETE) registry.105,106 The risk reported by Carrier et al. in a systematic review of RCTs and observational cohorts was 1.6% at 3 months.107 Therefore, the risk in the cohort reported by Kooiman appears to be similar to other cohorts receiving VTE treatment.104

The proportion of major bleeds that are fatal (25%) and the proportion of non-fatal major bleeds that are ICHs (9%) was based on data from the RIETE registry, which provides a detailed breakdown of major bleeds in patients receiving treatment for VTE.

Risk of chronic complications

The risk of CTEPH in patients having PE was taken from a systematic review by Ende-Verhaar et al.108 A cumulative incidence of 3.2% (95% CI 2.0 to 4.4) over 2 years was reported in those who survived the initial 3-month period after PE and this was applied to the model giving a risk of 1.6% per annum. We assumed no risk of new CTEPH beyond 2 years based on a study with a median follow-up of 94 months which reported no new cases of CTEPH after 2 years.109

The risk of PTS in patients with symptomatic DVT in the model was estimated from a prospective registry [Thrombosis and Pulmonary Embolism in Out-Patients (TULIPA) PLUS], reported by Hach-Wunderler et al.110 This source was chosen as the study design excluded patients with previous DVT or signs of PTS prior to the index DVT. The incidence is reported separately for proximal and distal DVT and is stratified into mild, moderate and severe PTS using a validated scoring system (Villalta scale), which is the scale preferred by the ISTH. The prevalence of symptomatic PTS at 3 years was 24.4% for all DVTs, 32.4% for proximal DVT and 15.6% for distal DVT, with 70% being mild, 24% being moderate and 6% being severe (average across both proximal and distal). Although the TULIPA registry recruited outpatients, 63.7% of the cohort were considered to have had provoked DVT, with 23.7% having had surgery in the last 12 weeks and 14.8% having had acute illness in the past 4 weeks. Given that we are assuming that DVT occurs post discharge and the majority of DVTs are managed as outpatients, it seemed reasonable to use data from this outpatient population in the current analysis.

Given that patients with asymptomatic DVT will remain undiagnosed and untreated, the incidence for asymptomatic proximal DVT was inflated using data from van Dongen et al. which reported an odds ratio of 2.71 (95% CI 1.44 to 5.10) for the risk of PTS in patients with symptomatic proximal DVT with good versus poor anticoagulation [defined as spending more than 50% of their time outside of the target international normalised ratio (INR) range].111 The rate of PTS in asymptomatic distal DVT was not inflated because this would have increased the incidence to a level similar to that seen in symptomatic proximal DVT and the clinical experts believed that the rate in untreated distal DVT should be lower than for treated proximal DVT. The timing of PTS was based on information from van Dongen et al., which reported the cumulative incidence over 4 years. This data was used to distribute the cumulative incidence reported by Hach-Wunderler et al. over 4 years. As van Dongen et al. reported no further incidence beyond 4 years we assumed no new incidence of PTS beyond 4 years in the model.

Mortality risks for adverse outcomes

Patients who survive the first 28 days after having an ICH have an increased risk of mortality in the long term. Folgelholm et al. report a standardised mortality ratio (SMR) of 4.5 in the first year and 2.2 in years two to six for patients who survived 28 days after ICH, compared to age- and sex-matched controls.112 We have applied the latter in years 2–6 of the state-transition model to capture the increased risk of death in the long term for patients having non-fatal ICH. However, the SMR reported for the first year for patients having non-fatal ICH is lower than the SMR of 9.4 for medical inpatients and 4.9 for surgical inpatients.83,85 Therefore, we have not applied any additional mortality risks for patients having non-fatal ICH in the first year as it did not seem logical that the mortality risks would be lower for those having ICH than for other medical or surgical inpatients not having ICH.

Two of the RCTs of LMWH versus placebo in medical inpatients reported the incidence of symptomatic PEs and the incidence of fatal PEs allowing the case-fatality rate to be calculated.88,90 Overall, across both the LMWH and the placebo arms, 29% (5/17) of the PEs that occurred were fatal PEs. This is substantially higher than the case-fatality rate reported in registries of patients presenting with VTE, such as the RIETE registry where the case-fatality rate was 3.2% overall, with a rate of 4.1% in the inpatient population and 2.9% in the outpatient population.113 However, given that the risk of death in the first 30 days reported by Moore et al. for all medical inpatients was 6.5%, it seems reasonable that the risk of death in those having PE would be higher than in all medical inpatients. It is also important to consider the fact that the RIETE registry only included patients diagnosed with PE and would miss cases where patients died suddenly and PE was only detected post-mortem. Studies that include PE diagnosed post-mortem have found a high rate of mortality even when excluding cases where autopsy discovered PE is classified as non-causal for death. Heit et al. report a 7-day survival rate of only 71.1% when using this approach, but a survival rate of 93% when excluding immediately fatal PEs or PEs first diagnosed at autopsy.114 This suggests that registries that recruit patients only after confirming PE using objective tests are likely to miss a substantial proportion of fatal PEs. Heit et al. also found that patients having a PE while in hospital had a higher risk of death than those having PE in a community setting, and those having surgery in the 90 days prior to their PE had a lower risk of death than those not having previous surgery. This suggests that we should expect higher case-fatality rates for inpatients compared to outpatients and higher case-fatality rates for medical inpatients compared to surgical inpatients. However, Heit et al. do not report an absolute case-fatality rate specifically for medical or surgical inpatients. An estimate of the case-fatality rate for PE for all surgical (6%) and medical patients (44.7%) is provided in CG92 (Table 4-3 of CG92) based on the average across all RCTs. These figures represent the ratio of fatal PE to symptomatic PE across multiple RCTs included in the systematic reviews that informed CG92. These figures were used in the model, in preference to rates from registry studies, to reflect the higher expected case-fatality rate for PEs in hospital inpatients, particularly in medical inpatients. However, the case-fatality rate for medical patients was re-estimated after excluding a single study which included sudden unexplained deaths in its calculation of fatal PEs, and which had a high rate of case-fatality (67%). Including this study was not considered appropriate in the calculation of the case-fatality rate as in the model the case-fatality rate is applied to confirmed PEs. After excluding this study, the average case-fatality rate across the remaining five RCTs was 26.8% and this was applied to medical inpatients in the base case.88,90,115–117 However, as there is considerable uncertainty regarding the appropriate case-fatality rate we have explored the range of estimates observed in the RCTs (13% to 67%) for medical inpatients in sensitivity analyses.88,118 We have also explored the impact of applying the higher average rate observed in medical inpatients (26.8%) to the surgical cohort.

Mortality risks in patients with CTEPH having either medical or surgical management were based on survival curves reported by Goodacre et al., which were estimated from an international prospective registry of patients with CTEPH.119 Deaths related to PE occurring within 6 months of PE are already accounted for in the model.120 For this reason, the hazard of death for patients with CTEPH is only applied from 1 year onwards, with PE-specific mortality applied up to 6 months and all-cause mortality applied from 6 months to 1 year. To ensure that the risk of death in the CTEPH group was not artificially low compared with the risk of death in the general population, general population mortality risks were applied whenever these were higher than the risk in the CTEPH population, based on the survival curves.

Cost of prophylaxis

We have assumed that pharmacological prophylaxis is LMWH. Although NG89 recommends that LMWH is given for a minimum of 7 days, a survey of 25 exemplar centres suggests that the majority of hospitals give LMWH for the duration of admission, which is typically less than 7 days.76 We have, therefore, assumed that LMWH is given only for the duration of admission.

The mean length of stay in admitted patients based on HES data is 5 days when excluding day cases (HES: Hospital Admitted Patient Care Activity 2018–19).84 We were unable to obtain an estimate of the mean length of stay specifically for medical inpatients, as the breakdown is given by main speciality and some specialities include both medical and surgical patients. However, we note that the mean length of stay is similar for general medical patients and general surgical patients, and these are the two most common categories, suggesting that it is reasonable to apply the same estimate for both medical and surgical patients. We also identified a study by Fowler et al. which reported that the pooled mean duration of stay for surgical patients (excluding day cases) was 4.8 days in 2015, suggesting that this estimate of 5 days is reasonable at least for surgical inpatients.81 We, therefore, assume that both medical and surgical patients receive 5 days of LMWH.

It is assumed that the lowest cost preparation of LMWH is prescribed. Therefore, drug costs for LMWH were based on the cost of dalteparin (5000 units given every 24 hours by subcutaneous injection) as this had a cost of £14.12 over 5 days, which was lower than the cost for equivalent doses of enoxaparin (£15.14 for 40 mg every 24 hours) or tinzaparin (£17.82 for 4500 for units every 24 hours). In the base-case analysis, we assumed that LMWH is only continued until discharge and, therefore, the treatment can be administered by a hospital nurse. We have assumed that each administration requires 2.5 minutes of time for a band 6 hospital nurse, with a unit cost of £47 per working hour, giving an administration cost of £9.79 over 5 days.78 Therefore, the total cost of prophylaxis is £23.91 for both medical and surgical inpatients.

We have also provided results for a scenario analysis in which LMWH is continued 2 days post-discharge to give a total of 7 days of LMWH, as this is recommended in NG89 as the minimum treatment duration. For this analysis, we have applied the costs for 2 days of outpatient administration of LMWH by adjusting the costs estimated by Menakaya et al. for outpatient LMWH.121 The costs estimated by Menakaya et al. included clinical time for counselling and blood taking (including a 5-day platelet count), pharmacy dispensing and initial administration. These estimates include the costs of administration by a district nurse in 4% of patients to reflect the fact that not all patients will be able to self-administer. The costs were adjusted to reflect the shorter duration of outpatient administration in this case (2 days vs. 46 days). In this scenario analysis the costs of administration are increased to £81.26 and the drug costs are increased to £19.76, such that the total cost of prophylaxis is £101.03.

In the strategies examining the use of prophylaxis based on RAMs, we have assumed that the risk assessment will require 5 minutes of time spent by a hospital consultant, which is a cost of £9.08 when applying a unit cost per hour of £109.78 This cost is applied to all strategies involving the use of a RAM, but not to the comparator strategies of prophylaxis for all and prophylaxis for none. A sensitivity analysis has also been conducted to explore whether the optimal strategy would be different if it could be assumed that the risk assessment would result in no additional cost, with the aim of exploring whether the cost of the risk assessment itself is a significant source of decision uncertainty.

Costs of treating VTE

The average time to VTE in medical inpatients reported by Barbar et al. was 42 days but this estimate is based on only 37 events.15 The Million Women study reported by Sweetland et al. shows that postoperative VTE risk in women peaks at 3 weeks although it remains raised up to 12 weeks and the average time to VTE is not reported.122 However, both these figures suggest that the majority of VTE occurs after discharge in both medical and surgical patients. Therefore, treatment costs for VTE are based on the same assumptions applied in a previous cost-effectiveness analysis of prophylaxis in patients having outpatient lower limb immobilisation. These include healthcare contacts for diagnosis, a short-stay inpatient admission where necessary, and 3 months of outpatient anticoagulant treatment. Resource use for diagnosis and admissions are summarised in Appendix 2, Table 26.

It is assumed that any VTE occurring prior to discharge, would extend hospital length of stay by up to 2 days and would, therefore, have a similar cost to that incurred by patients having VTE as an outpatient who require a short stay admission of up to 2 days. To determine whether the model is sensitive to uncertainties regarding the proportion of VTEs occurring pre-discharge and the likelihood that these will delay discharge, we have conducted a sensitivity analysis in which we assume that all VTEs occur pre-discharge and these VTEs result in no discharge delays.

Treatment for VTE is assumed to consist of 3 months of either a VKA (warfarin) or DOAC. Patients having treatment with a VKA are also assumed to have initial anticoagulation with LMWH until VKA treatment is established. It is assumed within the base-case model that 40% of patients will have a DOAC based on the usage of DOACs in the PREFER-VTE cohort from within those countries where DOACs had been launched.105 An additional scenario analysis has been conducted to examine the potential impact of increased DOAC prescribing in the years since its launch. The costs for administering LMWH are based on the costs reported by Menakaya, but these were adjusted to reflect the shorter duration of treatment (7 days compared with 42 days in Menakaya). Administration costs for oral anticoagulants include nine attendances at an anticoagulation service over 3 months for those receiving VKA, and a nurse-led telephone consultation at 10 days for DOACs. In addition, all patients are assumed to require an appointment with a consultant after 3 months to assess whether ongoing treatment is needed.

Costs of treating major bleeding

The cost over 90 days of fatal haemorrhagic stroke provided by Luengo-Fernandez et al. was uplifted to current prices and applied as the cost of fatal bleeds in the decision tree phase of the model.123,124 This paper also provided costs over 90 days for non-fatal haemorrhagic stroke stratified by the level of disability. A weighted average cost was calculated across non-disabling, moderately disabling and totally disabling haemorrhagic strokes. This was then uplifted to current prices and applied as the cost of non-fatal ICH in the decision tree phase of the model. Luengo-Fernandez et al. also report the average costs per annum from 90 days to 5 years post-stroke, but these are not reported separately for haemorrhagic stroke. The costs of GP care and emergency care are reported to be statistically significantly higher post-stroke compared to the year before stroke. In addition, they report the cost of residential care in patients not living in residential care prior to their stroke. The total post-acute (beyond 90 days) costs for primary care, emergency care and residential care were calculated and uplifted to current prices and applied in the state-transition phase of the model to those having non-fatal ICH. A pro rata cost is also applied to those having stroke more than 90 days before the end of the decision tree phase of the model.

The cost of non-fatal non-ICH major bleeds was assumed to be similar to the cost of GI bleeds. This was estimated based on a weighted average cost for non-elective inpatient and non-elective short-stay management of GI bleeds using NHS reference costs for bleeds requiring single, multiple or no interventions (HRG codes FD03A to FD03H).125 This was based on clinical expert advice that in surgical patients, the majority of major bleeds are distant from the site of surgery and, therefore, GI bleeding is a reasonable proxy for the cost of major bleeding even in surgical patients having non-GI surgery. In addition, the costs of GI bleeding are considered to be a reasonable proxy for the management of major bleeding in surgical patients, as the NHS reference cost data show that the majority of GI bleeds (87%) are managed without intervention, which is consistent with the data from XAMOS which shows that the majority of major bleeds in surgical patients are managed without a return to theatre.103

No costs are applied in the base-case analysis for CRNMB because it is assumed that these will be managed within the existing cost of inpatient care for those experiencing a CRNMB during inpatient prophylaxis. However, in the scenario analysis examining the use of LMWH for up to 7 days, a cost for ED attendance without admission is applied to a proportion of those having prophylaxis based on the ratio of CRNMBs to all major bleeds (2.05) reported in the PREFER-VTE registry.105

Costs of managing PTS and CTEPH

The management of PTS is assumed to involve one first and one follow-up vascular surgery outpatient appointment in the first year after diagnosis and two follow-up GP appointments every year thereafter. This is consistent with the assumption applied in a previous analysis.74 An alternative cost based on the burden estimated in a US cohort is considered in a sensitivity analysis.126

Drug costs for medical management of CTEPH were based on the costs used in CG92, which were uplifted to give a cost of £18,507 per year. The costs for medically managed patients are applied each year to those surviving with CTEPH. The proportion of patients having surgical management for CTEPH (59%) is based on data from Delcroix et al.120 The cost for surgical management of CTEPH is based on a weighted average of the reference costs for complex thoracic procedures (DZ02H/J/K) giving an average cost of £8175. A proportion of patients having surgical management (29%) are assumed to require medical treatment as a bridging therapy (average of 4.6 months). Including these costs brings the total cost in the first year to £10,237 for surgical management. No costs are applied beyond the first year for those managed surgically.

Utility values

In our previous modelling of VTE prevention in lower limb injury, we identified a systematic review of utility values in VTE by Lubberts et al.127 Forward citation searching was used to identify more recent systematic reviews on this topic and this identified a more recent review by Ghanima et al. but this yielded no more recent studies with relevant data.128 Further forwards and backwards citation searching from the review by Ghanima identified one paper by Utne et al. which reported EQ-5D outcomes by trial arm in patients being treated with either rivaroxaban or warfarin for DVT.129 Additional ad hoc searches using Google Scholar, which were limited to 2017 onwards, identified two more recent publications of health-related quality of life (HRQoL) data from the PREFER-VTE registry study, which had been previously used to obtain utility estimates in the model for lower limb injury.113,130 The new data from this paper were extracted and used to calculate updated utility multipliers for DVT and PE giving utility multipliers of 0.962 and 0.960 for the first 6 months after DVT and PE, respectively. The utility multipliers in the second 6 months were 1.00 and 0.99, respectively. For PE these are higher than the multipliers applied in the previous model because the utility values reported in the recent paper were much higher than those reported previously from the same study. This is because the final publication reported results separately for patients with and without active cancer and we selected the data for those without cancer for use in the model. However, it should be noted that the difference in utility between those with and without active cancer is fairly constant across time for PE, suggesting that the utility decrement attributable to PE would be similar, but that there is an additional decrement due to active cancer. In the previous model, the average utility for PE over 6 months was compared to the utility achieved 6 months after DVT to determine the multiplier on the assumption that patients with DVT will have returned to full utility by 6 months. Whereas in the analysis of the new data, the utility multiplier was compared against the norm value index score for an age-/gender-matched general population (0.838 for PE and 0.852 for DVT) provided in the more recent papers. This method is preferable as it is not affected by differences in the characteristics of those experiencing PE and DVT.

Using forward citation searching from the review by Lubberts et al. we identified a systematic review of studies reporting HRQoL after adverse events associated with antithrombotic therapy for primary or secondary prevention of VTE. However, only four studies were identified in this review and they related to VTE prevention in populations with atrial fibrillation or following acute myocardial infarction. Two of these studies reported EQ-5D in patients with and without a bleeding adverse event,131,132 but only one reported the EQ-5D for patients with major bleeding compared to no bleeding.132 After adjusting for baseline differences in EQ-5D, Amin et al. (2016) report that the EQ-5D decrement for major bleeding (BRAC bleeding type 2–4) in the past 6 months compared to no bleeding in the past 6 months was (−0.0445, 95% CI 0.073 to 0.016, p = 0.0022). As the utility in those without bleeding was 0.86 at both baseline and 6 months, this decrement is equivalent to a 5% proportional decrement. In our previous analysis for lower limb immobilisations, we assumed that the utility in the month following a GI bleed would be equivalent to the utility in the month following PE. Using data from the more recent publication of the PREFER-VTE registry130 would give a 16% decrement (0.718 compared to 0.838) when making this same assumption. The difficulty here is that the utility decrement associated with a major bleed is likely to reduce over time which is why we previously stated that any decrement would only persist for 1 month after a major bleed. Therefore, the study by Amin et al., which is focused on bleeding at any time in the past 6 months, may miss the maximum period of utility decrement. For this reason, we have chosen to make the same assumption as before and to apply the utility decrement in the first month post-PE to the first month post-major bleeding using data from the more recent PREFER-VTE publication.130

Utility values following ICH were based on data from 5-year follow-up of the Oxford Vascular Study (OXVASC) study as these data were previously applied in our analysis of prophylaxis during lower limb injury.123 A decrement of 0.22 was assumed in the decision tree part of the model where time since stroke was < 6 months and a decrement of 0.09 was assumed in the long-term part of the model. This study was chosen as the source of utility values previously as the duration of follow-up allowed time since stroke to be accounted for, and a comparison was made against general population norms.

In our previous analysis of prophylaxis during lower limb injury (Pandor et al. 201974), we used utility data from the CaVenT study to estimate the utility decrement in patients with PTS.133 An estimate of 10% was applied from diagnosis onwards, which was obtained from the CaVenT study by comparing the EQ-5D scores in those with and without PTS at 2 years. The CaVenT study did not stratify the utility estimates by severity of PTS, so this estimate was applied to all patients with PTS in the model regardless of severity. This may overestimate the utility decrement if the proportion of patients having severe PTS is lower in the modelled population than in the CaVenT study which recruited patients with acute iliofemoral DVT. A study by Lenert and Soetikno reported utility estimates for mild and severe PTS (0.98 and 0.93, respectively) obtained by using health state descriptions and a standard gamble valuation technique in a sample of volunteers.134 These were not used in the base case as utility measured using the EQ-5D in patients with the condition is preferable to utility measured using standard gamble in volunteers based on descriptions of the condition. However, a scenario analysis was conducted in which the data from Lenert and Soetikno were combined with data on the proportion of PTS that is severe (6%) from Hach-Wunderle et al. to estimate a utility decrement of 2% across all patients with PTS. In the current analysis we have taken the same approach and have applied a 10% decrement in the base case and a 2% decrement in a scenario analysis.

The utility decrement in patients with CTEPH was estimated from a study by Meads et al. by comparing the utility in patients having CTEPH (0.56) and the utility in patients with disease categorised as New York Heart Association (NYHA) class 1 (0.89) in which the quality-of-life impact of symptoms would be expected to be minimal.135 This gave a utility multiplier of 0.63 or a 37% decrement. This decrement is applied lifelong in the model to those having medical management of CTEPH, but only for 1 year in those having surgical management who have the utility multiplier for PE applied thereafter.

In our previous analysis of thromboprophylaxis during lower limb injury we identified several sources which estimated the utility decrement associated with VTE prophylaxis or VTE treatment.74 The study selected for use in the previous model was a study by Marchetti et al. which reported that patients would be willing to trade 2.7 of 365 days to avoid treatment with LMWH and 4 of 365 days to avoid treatment with warfarin.136 These data were previously used to estimate utility decrements of 0.007 for thromboprophylaxis and 0.011 for VTE treatment. These same decrements have been applied in this model.

Utility values for patients not experiencing any utility decrement due to prophylaxis, treatment, symptomatic VTE events, bleeding events (ICH or other major bleeds), long-term sequelae following VTE (PTS or CTEPH) or death are based on general population norms for a cohort of the same age and this is allowed to vary as the cohort ages during the model.137

Timing and duration of utility decrements applied in the decision tree

In order to calculate the QALYs gained by patients having different paths through the decision tree, it is necessary to make some assumptions regarding the timing of events as these are not explicitly modelled in a decision tree. We made the following assumptions when estimating QALYs in the decision tree:

• Baseline utilities using general population utility values for the starting age are applied to those not having treatment and not having any clinical events (e.g. VTE, bleeds).
• Bleeds during prophylaxis are assumed to occur halfway through the hospital admission (i.e. 2.5 days after starting prophylaxis).
• VTE is assumed to occur within 90 days of hospital admission and utility decrements for PE and DVT are applied from 42 days until the end of the decision tree period.
• Bleeds occurring during treatment for VTE are assumed to occur at 13, 32 and 12 days post diagnosis of VTE for fatal, ICH and other major bleeds, respectively (based on data from the RIETE registry reported by Nieto et al.).
• Disutilities for ICH are applied lifelong but separate disutilities are applied in decision tree and state-transition phases of the model (i.e. during the first 6 months after admission and beyond 6 months after admission).
• Disutility of non-fatal non-ICH major bleeding is assumed to last a maximum of 28 days.
• Disutility of prophylaxis applies for the duration of prophylaxis (5 days) and is therefore less in those stopping early due to major bleeding.
• Disutility of treatment for VTE applies for the duration of treatment (3 months) and is therefore less in those stopping early due to major bleeding.

Although the utility decrements for VTE are assumed to apply from 42 days onwards based on the mean time to VTE reported by Barbar et al.,15 there is likely to be considerable variability in the timing of VTE, which could occur before discharge or as late as 90 days after admission. Therefore, a sensitivity analysis was conducted in which the timing of VTE was varied from 3 to 90 days to determine if the model was sensitive to the assumption regarding the timing of VTE. We made the following assumptions when estimating QALYs in the state-transition model:

• Utility values for patents without any long-term sequelae (ICH, CTEPH, PTS) are taken from general population values and decrease as patients age in the model.
• All other utility values are applied as multipliers such that the absolute utility value decreases due to ageing in all patients.
• Utility decrements continue in the state-transition model for the remainder of the patient’s lifetime for PE but not for DVT where patients are assumed to return to general population utility values at 6 months.
• Patients with CTEPH who are treated medically have a lifelong utility decrement, whereas those treated surgically return after 1 year to the same utility as those surviving PE without CTEPH.
• Patients with PTS have the same utility decrement from diagnosis to death.
• Patients with ICH have the same utility decrement from 6 months (i.e. the start of state-transition model) to death.

Sensitivity and specificity of RAMs

Sensitivity and specificity data have been extracted from the RAM studies listed in Chapter 3.29,30,36,38,39,71 For many of the RAM studies we wished to examine in the model, the authors themselves have not reported sensitivity and specificity data. In this case, estimates of sensitivity and specificity have been extracted using reports of the incidence by risk score and graphical data, where possible. In some cases, estimates of sensitivity and specificity were extracted digitally from receiver operating characteristic (ROC) curves where these were reported.

Figure 7 shows the sensitivity and specificity data for RAMS in medical inpatients extracted from various studies. Overall, there is data reported for seven RAMs (Caprini, Kucher, Padua, IMPROVE, Intermountain, Geneva, Rothberg) and all except one (Rothberg) are reported in more than one cohort. It can be seen that there is not always consistency in the estimates of sensitivity and specificity for a particular RAM when reported across different cohorts. In particular, the estimates for IMPROVE and Intermountain obtained by Blondon et al. are more favourable than the estimates from Greene et al. However, several RAMs appear to have similar performance when estimates of sensitivity and specificity were taken from single cohort, as reported in papers by Greene et al. and Grant et al.

FIGURE 7 ROC curve for multiple risk assessment models evaluated in cohorts of medical inpatients. (Figure not reproduced here; see the original article.)

Figure 8 shows the sensitivity and specificity data for two RAMs in surgical inpatients. There were data on the Pannucci score reported by Pannucci et al. and data on the Caprini score reported by Bahl et al.

FIGURE 8 ROC curve for multiple risk assessment models evaluated in cohorts of surgical inpatients. (Figure not reproduced here; see the original article.)

Figure 9 shows the sensitivity and specificity data available for mixed populations of both surgical and medical inpatients. Only one paper reported this information for a mixed population (Elias et al. 201736). Elias et al. provide ROC curves for patients having prophylaxis and patients not having prophylaxis. They also provide a single estimate of sensitivity and specificity for the cohort as a whole.

FIGURE 9 ROC curve for the Padua RAM in a mixed cohort of medical and surgical patients reported by Elias et al.36 [with/without prophylaxis (PPX) and all]. (Figure not reproduced here; see the original article.)

The base-case deterministic analysis estimates the expected costs and QALYs for all available RAMs where performance data are available in that specific population (i.e. surgical or medical patients). In addition, because the usage of prophylaxis within the cohorts will tend to result in RAM performance being underestimated, we have also conducted a secondary analysis using the data from Elias et al. which was estimated in the subset of the cohort not receiving prophylaxis. This cohort included both surgical and medical inpatients and is therefore not used in the base-case scenario for either group.

Available data suggest that RAMs have generally weak predictive accuracy for VTE. There is insufficient evidence and too much heterogeneity to recommend one RAM over another. Hence, we have focused on using data from studies of RAMs to explore the trade-off between sensitivity and specificity, rather than try to identify the most cost-effective RAM. To do this we have used regression to estimate the trade-off between sensitivity and specificity for a typical RAM and have used this to explore how this trade-off affects the cost-effectiveness of using RAMs to select patients for prophylaxis.

Figure 10 shows the sensitivity and specificity data available across all populations with trend lines fitted firstly to all of the estimates across the multiple RAMS reported by Greene et al. and Grant et al. which were estimated in a single cohort, and secondly to the estimates from Elias et al., which were estimated in the subset of the cohort not having pharmacological prophylaxis. The trend lines have been fitted by assuming that there is a linear relationship between logit (sensitivity) and logit (1-specificity). It should be noted that for the line fitting against the data from Greene et al. and Grant et al., the fitted line assumes that each point is equally valid and does not take into account (1) the uncertainty around each pair of sensitivity and specificity estimates, (2) the relationship between multiple points that represent the ROC curve for a single RAM in a single study or (3) any relationship between estimates for different RAMs from the same study. The estimates from the Elias cohort were extracted for use in a sensitivity analysis exploring the potential for a more accurate RAM to be cost-effective. This was undertaken specifically to address the concern that use of VTE prophylaxis could lead to underestimation of the accuracy of the RAMs. It should be noted, however, that the Elias study reported higher accuracy for the Padua RAM than other studies in cohorts with and without prophylaxis. It is, therefore, unclear why accuracy was better in the Elias study, but it probably reflects features of the study design.

FIGURE 10 ROC curve for all studies (surgical, medical and mixed) with trend lines fitted. (Figure not reproduced here; see the original article.)

Model inputs for the elective hip replacement and elective knee replacement scenarios

In the main analysis for surgical inpatients, we excluded patients having elective hip and knee replacement surgery as the NICE guideline (NG89) recommends a more extended form of thromboprophylaxis in these groups. In knee replacement patients NG89 recommends LMWH for 14 days. In hip replacement patients NG89 recommends LMWH for 28 days. For comparison, in the majority of other categories of surgical patients, the duration of prophylaxis with LMWH recommended in NG89 is a minimum of 7 days.

If a single RAM is to be used across a broad range of surgical patients, encompassing populations that have a longer duration of prophylaxis, it is important to know whether the longer duration of prophylaxis in these groups would result in a different threshold for cost-effective use of prophylaxis. Therefore, we have conducted scenario analyses to determine what the risk threshold for thromboprophylaxis would be for elective hip and elective knee replacement patients and whether a prophylaxis for all strategies would be better in these specific populations.

In the scenario analyses for the two specific surgical populations of elective hip replacement patients and elective knee replacements, the overall modelling approach was the same as in the analysis for the general surgical cohort, but specific parameters were changed to make them more relevant to these groups. To identify relevant parameters, we used the cost-effectiveness models described for these populations in NG89. The key parameters for these populations are summarised in Table 6.

The absolute risks of VTE in elective hip and knee replacement patients were taken from the models in NG89. In NG89 these were estimated by combining information on the risk of symptomatic VTE in patients having prophylaxis taken from the National Joint Registry estimates of the ratio of asymptomatic to symptomatic DVTs provided by Quinlan et al. and information on the efficacy of prophylaxis taken from the NMAs conducted for NG89.99,100,138

The bleeding risk for elective hip and knee replacement patients has been taken from the model inputs for NG89. In the elective knee replacement population, the risk of major bleeding with LMWH was 1.23% (0.64% major bleeding at surgical site, 0.39% GI and ICH, 0.2% other major) in the NG89 model. For elective hip replacement patients, the baseline risk of major bleeding in those having standard duration LMWH was 3.21% in the NG89 model (2.29% major surgical site bleeding, 0.72% ICH and GI, and 0.2% other major). This rate of major bleeding was applied to elective hip replacement patients having extended duration LMWH in the model, as although the NMA in NG89 predicted a lower risk of major bleeding for extended LMWH versus standard duration LMWH (0.78, 95% CI 0.11 to 3.85), the difference was not statistically significant and applying a lower risk of bleeding for a longer duration of prophylaxis was considered to lack clinical face validity.

The proportion of major bleeds that were fatal or non-fatal ICHs was taken from the same source as used in the main surgical cohort as this was a study of patients having major orthopaedic surgery.103 Age at baseline and the proportion who are male were based on the estimates from the National Joint Registry used in NG89. The standardised mortality rates versus the general population were adjusted to give the same mortality rate at 12 months as applied in the NG89 model (1.49% for hip and 1.05% for knee), which again were based on data from the National Joint Registry. In hip replacement patients the case-fatality rate for PEs (17%) was taken from the model in NG89 and this was derived from the RCTs of elective hip replacement patients included in the reviews for NG89. However, in NG89 no equivalent data were available for knee replacement patients, so it was assumed that the case-fatality rate for knee replacement patients would be the same as for general surgical patients (6%).

The type of pharmacological prophylaxis used is assumed to be LMWH, but the duration of prophylaxis is assumed to be longer at 14 and 28 days, respectively, for knee and hip replacement patients compared with up to 7 days for the main surgical cohort. Also, because of the need for a longer duration of prophylaxis, it is assumed that prophylaxis will continue after discharge with outpatient prophylaxis used after discharge. For general surgical patients, in the scenario analysis in which we assumed that prophylaxis continues for 7 days, including 2 days of post-discharge prophylaxis, the total cost was £101.03. This total includes £67.55 for outpatient administration, which covers the cost of teaching patients to self-administer prophylaxis after discharge and the need for district nurse administration for a minority of patients (4%). In the elective knee replacement population, with a total of 14 days of LMWH, the total cost is increased to £141.71 due to the additional 7 days of LMWH. In the elective hip replacement population, with a total 28 days of LMWH, the total cost is £223.09.

NG89 also recommends prophylaxis with aspirin or DOACs as alternatives to LMWH in patients having elective hip or knee replacement. Therefore, we have conducted sensitivity analyses to see if adjusting the costs to reflect these alternative prophylaxis regimens, while assuming equivalent efficacy, results in different conclusions regarding the optimal use of RAMs in these cohorts. For the knee replacement population, the alternative regimens modelled consist or 14 days of rivaroxaban, at a cost of £25.20 or 14 days of aspirin, at a cost of £0.69. For the hip replacement population, the alternatives modelled consist of 35 days of rivaroxaban, at a cost of £63.00, or 10 days of LMWH followed by 28 days of aspirin, at a cost of £119.84.

Economic modelling results

Clinical outcomes predicted by the model with and without prophylaxis

Table 7 shows the clinical outcomes predicted by the model with and without prophylaxis in each of the modelled populations.

In medical inpatients, prophylaxis reduces serious adverse outcomes (fatal PEs, fatal bleeds and non-fatal ICHs) from 53 per 10,000 to 42 per 10,000. The reduction in symptomatic DVTs and non-fatal PEs is also higher than the increase in other bleed types. In the long-term outcomes (presented in Table 7) at 5 years, there is a large reduction in the number of patients experiencing PTS from prophylaxis from 787 per 10,000 to 385 per 10,000, but minimal difference in overall survival. Therefore, in medical inpatients, although prophylaxis increases the risk of major bleeding, this is more than outweighed by the number of VTE events avoided and there are additional benefits from preventing long-term VTE-related complications such as PTS.

In surgical inpatients, the risk of a serious adverse outcome (fatal PEs, fatal bleeds and ICH) is low at 7 per 10,000 but it is increased slightly by prophylaxis to 11 per 10,000 due to the additional risk of fatal bleeding or non-fatal ICH. Although the risk of any symptomatic VTE is reduced from 140 per 10,000 to 41 per 10,000, the risk of any major bleeding is increased from 125 per 10,000 to 370 per 10,000. Prophylaxis still reduces the risk of PTS in surgical inpatients, from 367 per 10,000 to 107 per 10,000, but PTS is less common in the surgical cohort than in the medical cohort due to the lower risk of DVT. Therefore, in the surgical population there is a more explicit trade-off between the short-term risk of major bleeding and the benefits of preventing VTEs and their long-term consequence.

In elective hip replacement patients, the risk of a serious adverse clinical outcome (fatal PEs, fatal bleeds and non-fatal ICH) is reduced by prophylaxis from 28 per 10,000 to 17 per 10,000. The risk of any symptomatic VTE is reduced from 346 per 10,000 to 116 per 10,000 and the risk of any major bleeding is increased from 128 per 10,000 to 321 per 10,000. The reduction in PTS is similar to that observed in the general surgical inpatient population (294 per 10,000 reduced to 98 per 10,000 by prophylaxis). Therefore, in the elective hip replacement population, the increase in rates of major bleeding is similar to the decrease in rates of symptomatic VTE, but overall there is a reduction in serious adverse clinical outcomes as only a very small minority of major bleeds are fatal or intracranial.

In elective knee replacement patients, the risk of a serious adverse clinical outcome (fatal PEs, fatal bleeds and non-fatal ICH) is reduced from 20 per 10,000 to 11 per 10,000 by prophylaxis. The risk of any symptomatic VTE is reduced from 397 per 10,000 to 186 per 10,000 but the increase in the risk of any major bleeding is smaller (116 per 10,000 to 125 per 10,000). The reduction in PTS is more similar to that observed in the medical inpatient population (764 per 10,000 reduced to 357 per 10,000), which is largely being driven by the high risk of asymptomatic DVT in the elective knee replacement population which is more similar to that observed in the medical inpatient population.

The 5-year survival is higher in surgical inpatients than in medical inpatients which is to be expected given the lower starting age of 54 years in surgical inpatients compared with 64 years in medical inpatients. In both the elective knee replacement population and the elective hip replacement population, all-cause mortality at 5 years is higher than in the general surgical population due to a higher starting age of 69 years in both groups. However, the impact of prophylaxis on all-cause mortality at 5 years is small regardless of the population selected.

Balance of benefits and harms of prophylaxis at different levels of risk

It can be seen from the clinical outcomes predicted by the model in Table 7, that there is always a trade-off between the benefits of using prophylaxis to reduce VTE risks and the potential harms that may occur due to the increased risk of bleeding. These differ between the medical and surgical populations largely because these groups have different average risks of VTE and major bleeding but also because some specific types of VTE outcomes and major bleeding events are more likely in some populations than others. It would, therefore, be useful to explore the overall impact of prophylaxis on health at differing levels of risk so that the balance of benefits and harms in individuals who have a higher or lower than average risk can be assessed. Figures 11 and 12 present information on whether the overall impact of prophylaxis on QALYs over the patient’s remaining lifetime is positive or negative for different levels of VTE risks and major bleeding risks. Where the overall impact of prophylaxis on lifetime QALYs is positive for a given level of VTE and major bleeding risk, the cell is coloured green and where it is negative the cell is coloured red. It should be noted that the steps in risks between rows and columns are not even across the table. In addition, although the risk levels presented for VTE without prophylaxis, and for major bleeding with prophylaxis are the same in Figures 11 and 12, the risk levels for VTE with prophylaxis and major bleeding without prophylaxis differ because the RRs for these outcomes are different in medical and surgical inpatients. The square outlined in bold is the average level of risk assumed for that population in the base-case analysis.

FIGURE 11 Balance of benefits and harms for medical inpatients with different levels of risk (blue and red indicate overall QALY gain and loss, respectively). (Figure not reproduced here; see the original article.)

FIGURE 12 Balance of benefits and harms for surgical inpatients with different levels of risk (blue and red indicate overall QALY gain and loss, respectively). (Figure not reproduced here; see the original article.)

In Figure 11, it can be seen that prophylaxis for all results in QALY gains for medical inpatients with average levels of VTE and bleeding risks (3.4% and 1.02%, respectively). Prophylaxis for none would result in greater QALY gains if either the average VTE risk in the medical cohort was much lower (≤ 0.35%) or the average risk of bleeding in the medical cohort was much higher (≥ 11%). However, in a cohort of patients with both a higher risk of bleeding and a lower risk of VTE, for example, a 6% bleeding risk and a 1.5% VTE risk, prophylaxis would also do more harm than good.

In Figure 12, it can be seen that prophylaxis results in overall QALY gains over a much broader range of VTE and bleeding risks in surgical inpatients than in medical inpatients. In surgical inpatients a much lower proportion of major bleeding is fatal bleeding or a non-fatal ICH and, therefore, the overall impact of prophylaxis is positive over a broader range of bleeding risks. In addition, the average age assumed for patients in the surgical population is lower than the average age for patients assumed in the medical population. Therefore, any QALY losses due to long-term consequences of VTE, such as PTS, will be larger as they will have an impact over a longer time frame.

Figures 11 and 12 show exploratory analyses and should not be used to guide decision-making. The analysis assumes that the RR of bleeding with prophylaxis is constant and does not increase as the baseline bleeding risk increases, whereas it is conceivable that risk factors for bleeding also increase the RR of bleeding with prophylaxis. Furthermore, VTE and bleeding risk factors are often inter-related and there may be interactions between risk factors and between risks. These issues mean that the decision to prescribe prophylaxis for an individual patient with an increased bleeding risk is complex and unlikely to lend itself to determination by a simple trade-off.

Figures 11 and 12 show the balance of benefits and harms for prophylaxis in terms of the impact on overall lifetime QALYs, but they do not address the question of whether it is cost-effective to use prophylaxis at different levels of risk and whether it is cost-effective to use a RAM to select patients at higher risk for prophylaxis. These questions are addressed in Deterministic base-case cost-effectiveness results for medical inpatients to Deterministic scenario analyses for elective hip and knee replacement populations below.

Deterministic base-case cost-effectiveness results for medical inpatients

The deterministic base-case results obtained when applying the midpoint parameter estimates to the base-case scenario for medical inpatients using data for the five RAMS that were evaluated in a single cohort of patients (Green et al. 2016; Grant et al. 2016) are presented in Figure 13. For each RAM, several points are plotted which represent the costs and QALYs that can be achieved for different RAM cut-off points. It can be seen in Figure 13 that the strategy of prophylaxis for all dominates all other options because it has a larger expected QALY gain and it also has a lower expected cost. Figure 13 also shows the expected costs and QALYs for a RAM which has the average characteristics of all the RAMS evaluated in the cohort reported by Greene et al. and Grant et al. using the ROC curve regression described in Sensitivity and specificity of RAMs.

FIGURE 13 Deterministic base-case results for medical inpatients when using RAMs evaluated in a single cohort of patients (Greene et al.39 and Grant et al.38). (Figure not reproduced here; see the original article.)

Figure 14 shows the secondary analysis introducing the data from RAMs evaluated in other cohorts. For some of these RAMs, only a single point is plotted as sensitive and specificity was only reported at a single threshold but for others multiple points are plotted showing different RAM thresholds. Several of these RAMs appear to perform better than the regression based on the average RAM evaluated in the cohort reported by Greene et al. and Grant et al. in that they provide a similar number of QALYs at a lower cost. The best-performing RAM reported in an alternative cohort, is the Geneva score with a cut-off of ≥ 3 (sensitivity of 94.6% and specificity of 44%) reported by Blondon et al.29 However, prophylaxis for all would still be more cost-effective than using a Geneva score cut-off of ≥ 3, even when applying the performance data from Blondon et al.

FIGURE 14 Deterministic base-case results for medical inpatients when using RAMs evaluated by Greene et al.39 and Grant et al.38 and RAMs evaluated in other cohorts (including the mixed cohort of surgical and medical patients reported by Elias et al.36). (Figure not reproduced here; see the original article.)

Also shown in Figure 14 is a sensitivity analysis using the Padua RAM performance data from Elias et al. which was obtained in the subgroup of the cohort not having prophylaxis.36 When including this data, the optimal strategy would be a Padua score of ≥ 3 (sensitivity 99.9% and specificity of 23.7%), as this has slightly higher QALYs and slightly lower costs than prophylaxis for all. However, the caveat here is that Elias et al. recruited a mixed population of medical and surgical patients and, therefore, it is unclear if this RAM would perform equally well in an exclusively medical cohort. When the Padua score was evaluated by Greene et al. in an exclusively medical cohort, the performance was not as good, with a Padua score of ≥ 3 having a sensitivity and specificity of 49.3% and 73.0%, respectively. However, this sensitivity analysis suggests that if a RAM with sufficiently good performance could be developed and validated in an exclusively medical cohort, then prophylaxis for all would not be the optimal strategy in medical patients. Also shown are the expected costs and QALYs when using a regression fitted to the data points from Elias et al. It can be seen from this that the performance based on the regression is not as good as when using the observed sensitivity and specificity for the actual Padua RAM scores obtained by Elias et al. This is because the regression is not a good fit to the ROC curve for low RAM scores.

Probabilistic base-case results for medical inpatients

Based on these deterministic analyses, the probabilistic model was run for the strategies of no prophylaxis, prophylaxis for all and for the various cut-offs of the Padua score (assuming the RAM performance observed in Greene et al.). It should be noted that this probabilistic analysis does not explore the uncertainty regarding which RAM is optimal as this would require comparing all of the cut-offs from all of the various RAMs simultaneously. Instead, it uses the Padua RAM as an example RAM to answer the question of what the optimal strategy would be if the RAM chosen performed similarly to the Padua RAM.

The results based on the mean outcomes from 10,000 probabilistic model runs are summarised in Table 8. The broad conclusions are the same, in that prophylaxis for all is the optimal strategy when applying valuing a QALY at £20,000–30,000.

Figure 15 shows that while the average cost saving is £28.44 and the average QALY gain is 0.055 QALYs for prophylaxis for all versus prophylaxis for none, there are a broad range of estimates with 79% of samples resulting in a cost saving and the incremental cost having a CrI of £−113.66 to £46.28. The incremental QALY estimate has a CrI of 0.02 to 0.11, with 99.9% of samples resulting in a QALY gain compared to prophylaxis for none.

FIGURE 15 Cost-effectiveness plane for prophylaxis for all versus prophylaxis for none in medical inpatients. (Figure not reproduced here; see the original article.)

The cost-effectiveness acceptability curve (CEAC) is presented in Figure 16 for medical inpatients. There is < 1% chance that prophylaxis for all is not the optimal strategy when valuing a QALY at £20,000. In addition, because prophylaxis for all is cost-saving compared to even the next most effective strategy, there is a 76% chance that it would be the optimal strategy when valuing a QALY at £0.

FIGURE 16 Cost-effectiveness acceptability curve (CEAC) for medical inpatients (applying performance of Padua RAM from Greene et al.39). Strategies with < 1% probability across all values of a QALY are not plotted. (Figure not reproduced here; see the original article.)

The overall expected value of perfect information (EVPI) associated with all parameters included in the probabilistic sensitivity analysis (PSA) when valuing a QALY at £20,000 was £0.06 per patient per year (£59.33 per 1000 patients per year) due to the high probability that prophylaxis for all is the optimal strategy. This suggests that there is minimal decision uncertainty regarding whether a RAM should be used in medical inpatients, when assuming the RAM will perform according to the performance of the Padua RAM in the Greene et al. cohort.

However, the results in Figure 14 suggest that the optimal strategy may not be prophylaxis for all if a better-performing RAM could be developed in medical patients. Therefore, we decided to re-run the EVPI analysis for the medical cohort when using the sensitivity and specificity data from the mixed cohort of medical and surgical patients reported by Elias et al. When using this sensitivity and specific data for medical inpatients, the EVPI is £2.42 per person per year (£2422.90 per 1000 patients per year). Offering prophylaxis at a RAM score of ≥ 3 has a 76.6% chance of being optimal and prophylaxis for all has a 17.6% chance of being optimal when valuing a QALY at £20,000. In addition, a large proportion of the global EVPI is attributable to the uncertainty around the RR of bleeding for prophylaxis versus no prophylaxis; expected value of partial perfect information (EVPPI) of £2.01 for this parameter with no other parameter having an EVPPI > £0.01. This sensitivity analysis suggests that if a better-performing RAM could be developed for medical inpatients, then the decision uncertainty would largely relate to the uncertainty around the increased risk of bleeding attributable to prophylaxis.

Deterministic scenario and sensitivity analyses for medical inpatients

Deterministic scenario and sensitivity analyses were conducted to explore which model assumptions and inputs were key drivers of decision uncertainty. To do this, deterministic results were generated using midpoint parameter inputs when varying individual model inputs or assumptions. These analyses assume that the RAM used is Padua and that it performs as observed in the cohort reported by Greene et al.

The results presented in Table 9 show that the broad conclusion that prophylaxis for all is the optimal strategy (defined as most cost-effective when valuing a QALY at £20,000) does not vary under most of the scenarios and sensitivity analyses considered. The exceptions were scenarios that explored large changes to the risk of VTE and bleeding during PPX, with a sixfold decrease in VTE risk or a sixfold increase in risk of major bleeding needed to alter the optimal strategy.

On the basis of this, further analyses were conducted to explore how the optimal strategy varies when changing the average risks of VTE and major bleeding in the medical cohort. Figure 17 shows that the optimum strategy is prophylaxis for all when the risk of VTE is high and the bleeding risk is low, but the optimum strategy is pharmacological prophylaxis for none when the risk of VTE is low and the bleeding risk is high. There is a band of combined VTE and bleeding risks where using a RAM to target prophylaxis at higher-risk patients would be optimal instead of a blanket strategy of either prophylaxis for all or prophylaxis for none. This suggests that it is important to gather good evidence on the average risks of VTE and major bleeding in the medical cohort as a whole in order to determine whether a RAM should be used to direct prophylaxis.

FIGURE 17 Optimal risk assessment strategy for cohorts of medical inpatients with varying levels of baseline risks (box indicates base-case scenario). It should be noted that the rows and columns do not represent equal steps in risk; red (10) indicates prophylaxis for none is optimal and blue (0) indicates prophylaxis for all is optimal; the numbers in cells of other colours represent the optimal Padua RAM score at which prophylaxis should be offered at that level of risk to maximise incremental net monetary benefit when valuing a QALY at £20,000. (Figure not reproduced here; see the original article.)

The sensitivity and scenario analyses also suggest that various individual factors about which there was some uncertainty have no impact on the optimal strategy. These include the incidence of PTS after asymptomatic DVT, the costs and utility losses attributable to PTS, the case-fatality rate for PE and the exact timing of VTE within the 90-day post-admission period. In addition, it was found that the optimal strategy did not change when assuming no additional cost for administering the RAM or when assuming that patients continue LMWH for 7 days including 2 days of post-discharge administration.

Deterministic base-case results for surgical inpatients

The deterministic base-case results obtained when applying the midpoint parameter estimates to the base-case scenario for surgical inpatients using data for the two RAMS that were evaluated in either a surgical population (Caprini by Bahl et al.,26 Pannucci by Pannucci et al.54) are presented in Figure 18. For each RAM, several points are plotted which represent the costs and QALYs that can be achieved for different RAM cut-off points. For the Pannucci RAM, the cut-offs shown are ≥ 6, ≥ 3 and ≥ 1. For the Caprini score the cut-offs shown are ≥ 7, ≥ 5, ≥ 3 and ≥ 2. From this it can be seen that the optimal strategy is pharmacological prophylaxis for all, if the analysis is restricted to RAMs evaluated exclusively in this cohort. Figure 18 also includes results for the Padua RAM from the mixed cohort of both surgical and medical inpatients reported by Elias et al. When including the performance data from the Padua score evaluated by Elias et al., the optimal strategy is a Padua score of ≥ 3 (sensitivity of 99.9% and specificity of 23.7%), which achieves more QALYs at a reduced cost compared to prophylaxis for all.

FIGURE 18 Deterministic base-case results for surgical inpatients when using RAMS evaluated in either surgical populations (Pannucci et al. 201454, Bahl et al. 201026) or mixed populations of both surgical and medical inpatients [Elias without prophylaxis (PPX)]. (Figure not reproduced here; see the original article.)

Figure 18 also shows the expected cost-effectiveness of a RAM whose performance is based on a regression fitted to the data from the ROC curve for Padua using the data observed in the cohort not having thromboprophylaxis reported Elias et al. The data for the lower Padua score cut-offs (≥ 1 to ≥ 3) were excluded from this regression as they did not fit the assumption that there is a linear relationship between logit (sensitivity) and logit (1-specificity). Therefore, the regression does not closely follow the performance of the Padua score for the cut-offs below 4 and the Padua score is predicted by the regression to perform less well than observed at cut-offs below 4. This results in thromboprophylaxis for all being the optimal strategy when assuming that the RAM would follow the ROC curve predicted by the regression. Based on this, we can say that a RAM would need to have a very high sensitivity, similar to that observed by Elias et al., to make using a RAM more cost-effective than offering prophylaxis to all.

Probabilistic base-case results for surgical inpatients

Based on these deterministic analyses, the probabilistic model was run for the strategies of no pharmacological thromboprophylaxis, prophylaxis for all and for the various cut-offs of the Pannucci score. It should be noted that this does not capture uncertainty related to which RAM is optimal as this would require a simultaneous comparison of all of the various cut-offs for all three RAMs.

The results based on the mean outcomes from 10,000 probabilistic model runs are summarised in Table 10. The broad conclusions are the same, in that offering prophylaxis to surgical patients with a Pannucci score ≥ 3 is the optimal strategy when applying valuing a QALY at £20,000 to £30,000.

Figure 19 shows the incremental costs and QALYs for pharmacological thromboprophylaxis for all versus prophylaxis for none in surgical inpatients. It can be seen that while the average cost saving is £47.88 and the average QALY gain is 0.035 QALYs, there are a broad range of estimates with 24% of PSA samples resulting in a cost saving and the incremental cost having a CrI of −£96 to £254. The incremental QALY estimate has a CrI of 0.002 to 0.080 with 98% of samples resulting in a QALY gain compared to prophylaxis for none.

FIGURE 19 Cost-effectiveness plane for prophylaxis for all versus prophylaxis for none in surgical inpatients. (Figure not reproduced here; see the original article.)

The CEAC is presented in Figure 20 for surgical inpatients. It can be seen from this that giving prophylaxis to all patients has the highest probability of being cost-effective at £20,000 per QALY, with a 70% likelihood of being optimal, but there is also a 17% likelihood that using a Pannucci score of ≥ 3 would be optimal and a 9% likelihood that using a Pannucci score of ≥ 1 would be optimal.

FIGURE 20 CEAC for surgical inpatients (assuming the RAM used is Pannucci). (Figure not reproduced here; see the original article.)

The overall EVPI associated with all parameters included in the PSA when valuing a QALY at £20,000 was £16.35 per person. This is calculated across all strategies including no prophylaxis, Pannucci scores of ≥ 6, ≥ 3 and ≥ 1 and thromboprophylaxis for all. It does not include the uncertainty of considering which RAM to use as the EVPPI has been run assuming that the Pannucci score is used. In addition, it should be noted that the EVPI analysis does not include any uncertainty not captured in the PSA such as uncertainty around the model assumptions or the choice of data sources, or uncertainty around the estimates of sensitivity and specificity.

In the analysis of parameter EVPI (EVPPI), only six individual parameters were found to have an EVPPI of more than £0.01 per person (see Table 11 EVPI results for surgical in patients when using Pannucci RAM based on sensitivity and specificity from Pannucci et al.). The parameter with the largest EVPPI was the RR of VTE for patients having thromboprophylaxis compared to patients not having thromboprophylaxis. Other slightly less important parameters were the likelihood that major bleeding is intracranial (ICH) or fatal, the absolute and RRs of bleeding and the utility of patients experiencing PTS following DVT. These findings reinforce the idea that it is the trade-off between the short-term risks of bleeding in patients receiving prophylaxis and the long-term risks of chronic complications that are most important to the decision regarding which surgical inpatients should be offered prophylaxis.

We also ran the PSA and EVPI analysis assuming that the Padua RAM is used and that it performs as seen in subset of the cohort reported by Elias et al. who did not receive prophylaxis. This was to explore whether further research would be more worthwhile if a RAM could be identified that performed better than the Pannucci RAM in surgical inpatients. The CEAC when assuming the RAM performance for Padua reported by Elias et al. is shown in Figure 21. When using the sensitivity and specificity data from Elias et al. the overall EVPI was £15.90 and the same set of parameters was identified as being the most important drivers of uncertainty.

FIGURE 21 CEAC for surgical inpatients (assuming the RAM used is Padua from Ellias et al.36). (Figure not reproduced here; see the original article.)

Deterministic scenario and sensitivity analyses for surgical inpatients

Deterministic scenario and sensitivity analyses were conducted to determine whether the optimal strategy (when valuing a QALY at £20,000) varies under different assumptions or alternative sources of evidence. These analyses assume that the RAM used is Pannucci. The optimal strategy is defined as the most cost-effective strategy when valuing a QALY at £20,000.

These analyses (see Table 12) identified that prophylaxis for all is no longer the optimal strategy when the risk of VTE is halved, or when the risk of bleeding is doubled. It is also no longer optimal when the utility decrement of PTS is halved or estimated from alternative sources,134 or when assuming there is no risk of PTS in asymptomatic DVTs. It is also no longer optimal when assuming that there are no asymptomatic DVTs. These analyses suggest that the cost-effectiveness of offering prophylaxis to surgical patients is being partly driven by the prevention of PTS in patients having asymptomatic DVT. The scenario exploring an increase in age to 80 changed the optimal strategy to a RAM score of ≥ 3, which suggests that the optimal strategy may differ for patients who have lower life expectancy and lower quality of life prior to admission as these patients have less potential to benefit from reducing the risk of long-term complications. An increase in the duration of prophylaxis to 7 days to cover 5 days of admission and 2 days of post-discharge prophylaxis also altered the optimal strategy to offering prophylaxis at a RAM score of ≥ 1. However, this was largely attributable to the additional 2 days of prophylaxis provided after discharge, as the duration of in-hospital prophylaxis had to be increased to 16 days before the optimal strategy changed. The use of extended prophylaxis for 28 days also resulted in an optimal strategy of offering prophylaxis for those with a RAM score ≥ 3.

The sensitivity and scenario analyses also suggest that various individual factors about which there was some uncertainty have no impact on the optimal strategy. These include the costs attributable to PTS, the case-fatality rate for PE and the exact timing of VTE within the 90-day post-admission period.

As these one-way scenario analyses identified that the conclusions were sensitive to the average risks of VTE and major bleeding in the cohort, we conducted a two-way scenario analysis to determine whether the optimal strategy would be different if the risk of VTE or the risk of major bleeding had been underestimated or overestimated in the base-case analysis. Figure 22 shows, although prophylaxis for all is the optimal strategy at the exact level of VTE and bleed risk assumed in the base-case scenario, the use of a RAM would be optimal across a large range of VTE and bleeding risks if that RAM performed similarly to the Pannucci RAM. This suggests that obtaining good evidence on the average baseline risks of VTE and major bleeding in the surgical cohort being offered risk assessment is important for determining which patients should be offered thromboprophylaxis.

FIGURE 22 Optimal risk assessment strategy for cohorts of surgical inpatients with varying levels of baseline risks (box indicates base-case scenario). It should be noted that the rows and columns do not represent equal steps in risk; red (10) indicates prophylaxis for none is optimal and blue (0) indicates prophylaxis for all is optimal; the numbers in cells of other colours represent the optimal Pannucci RAM score at which prophylaxis should be offered at that level of risk to maximise incremental net monetary benefit when valuing a QALY at £20,000. (Figure not reproduced here; see the original article.)

Deterministic scenario analyses for elective hip and knee replacement populations

Figure 23 shows the incremental costs and QALYs versus no prophylaxis for each RAM cut-off point. For each series of points, the point closest to the origin shows the costs and QALYs for a Pannucci ≥ 6, and then each successive point shows the next RAM cut-off (≥ 3, ≥ 1 etc.) until the last point shows prophylaxis for all. Separate lines show the results for the main surgical population and for the two specific populations of patients having either elective hip replacement or elective knee replacement. For the main surgical population, two alternative assumptions regarding whether thromboprophylaxis is stopped at discharge or continued post-discharge are shown for comparison.

FIGURE 23 Incremental costs and QALYs versus no prophylaxis for various RAM cut-off points (Pannucci ≥ 6, Pannucci ≥ 3, Pannucci ≥ 1, PPX for all) for populations requiring PPX after discharge. (Figure not reproduced here; see the original article.)

The optimal strategy in the base case for the main surgical population, where LMWH is assumed to stop on discharge at 5 days, is prophylaxis for all. This is also the optimal strategy in the knee replacement population where LMWH is assumed to continue post discharge up to 14 days. For the main surgical population, when allowing patients to continue prophylaxis post discharge up to a maximum of 7 days, the optimal strategy (when valuing a QALY at £20,000) is offering prophylaxis using a RAM score cut-off of ≥ 1. The optimal strategy in the hip replacement population where LMWH is assumed to continue post discharge up to 28 days is offering prophylaxis using a RAM score cut-off of ≥ 3.

When applying the costs for the alternative pharmacological prophylaxis strategies using either aspirin or DOACs, but assuming equivalent efficacy to LMWH, the incremental costs were lower than the base-case scenario as these thromboprophylaxis strategies had lower costs. In the case of elective knee replacement patients, the optimal strategy was thromboprophylaxis for all when assuming either 14 days of aspirin or 14 days of a DOAC were offered. In hip replacement patients, thromboprophylaxis for all was optimal when assuming that DOACs were offered instead of LMWH and offering prophylaxis at a Pannucci score of ≥ 1 was optimal when assuming that prophylaxis consisted of 10 days of LMWH was followed by 28 days of aspirin.

The additional costs of offering extended duration prophylaxis, particularly with LMWH instead of DOACs or aspirin, mean that using a RAM to reduce the population being offered prophylaxis is optimal in patients having an elective hip replacement. However, in patients having elective knee replacement, when the duration of prophylaxis is only 14 days, the optimal strategy is the same as in the main surgical cohort where prophylaxis is recommended up to 7 days.

Summary of key findings

• For medical inpatients, pharmacological prophylaxis for all patients without contraindications is likely to be the more cost-effective strategy, when valuing a QALY at £20,000. This assumes that any RAM applied in medical inpatients would perform in a similar manner to the average RAM performance estimated in a single cohort of medical inpatients.
• The optimal prophylaxis strategy was fairly robust under sensitivity analysis in the medical inpatient cohort. The key factors determining cost-effectiveness appear to be the baseline risks of VTE and major bleeding and the increased risk of major bleeding associated with prophylaxis.
• For surgical inpatients, pharmacological prophylaxis for all patients without contraindications is likely to be the more cost-effective strategy, when valuing a QALY at £20,000 and assuming that any RAM applied in a surgical population performs similarly to either the Pannucci or Caprini RAM.
• In the surgical inpatient cohort, the most cost-effective strategy was sensitive to many of the individual model inputs and assumptions tested, with the optimal strategy being a Pannucci score of ≥ 6 (sensitivity of 40% and specificity of 82%) when assuming no risk of PTS after asymptomatic proximal or distal DVT.
• In specific surgical populations who require longer durations of prophylaxis, it was found that pharmacological prophylaxis for all eligible patients is optimal in those having 14 days of LMWH following elective knee replacement whereas offering prophylaxis at a Pannucci score of ≥ 3 (sensitivity of 84% and specificity of 49%) was optimal when offering 28 days of LMWH to patients having elective hip replacement.
• Using a RAM to select patients for pharmacological prophylaxis in medical or surgical inpatients would only be optimal if that RAM had a high sensitivity, similar to that observed for the Padua RAM with a cut-off of ≥ 3 in a mixed cohort of medical and surgical patients (sensitivity 99.9%, specificity 23.7%).