Cost-effectiveness analysis of hydrophilic-coated catheters in long-term intermittent catheter users in the UK

Abstract

Objective

To estimate the cost-effectiveness of single-use hydrophilic-coated intermittent catheters (HCICs) versus single-use uncoated intermittent catheters (UICs) for urinary catheterization.

Methods

The evaluation took a UK national health service (NHS) perspective. The population of interest were people using intermittent catheters, with either a spinal cord injury or multiple sclerosis. A Markov model was developed that estimated costs and clinical evidence over the lifetime of a hypothetical cohort and applied health-related quality-of-life estimates. Model inputs were sourced from published evidence, including a network meta-analysis to inform the treatment effect (reduction in catheter-associated urinary tract infections [CAUTIs]), and were supported by expert opinion. The model outputs included per-patient lifetime costs, quality-adjusted life years (QALYs), and the incremental cost effectiveness ratio (ICER). Event counts were also produced.

Results

Using HCICs instead of UICs could prevent seven CAUTI events per patient over a lifetime horizon (1.8 requiring secondary care). Overall, lifetime use of HCICs is £3,183 more expensive than use of UICs per patient. However, for these additional costs, 0.55 QALYs are gained. The ICER is £5,755 per additional QALY gained. Key drivers of the model results were identified and subject to sensitivity analyses. The results were found to be robust to parameter uncertainty.

Conclusion

HCICs are likely to be a cost-effective alternative to UICs, a result driven by substantial reductions in the number of CAUTIs. Their adoption across clinical practice could avoid a substantial number of infections, freeing up resources in the NHS and reducing antibiotic use in urinary catheter users.

PLAIN LANGUAGE SUMMARY

A new economic model was developed from a United Kingdom National Health Service perspective, to explore whether hydrophilic-coated intermittent catheters would be “worth” introducing for intermittent catheter users with either a spinal cord injury or multiple sclerosis. More specifically, costs were analyzed alongside clinical evidence and health-related quality-of-life data to investigate whether hydrophilic-coated intermittent catheters would offer a notable health benefit when compared with uncoated intermittent catheters for the assessed population, whilst keeping costs to the National Health Service sufficiently low. Model inputs were sourced from published evidence where possible, and experts were consulted otherwise. The results showed that, whilst lifetime use of hydrophilic-coated intermittent catheters is £3,183 more expensive than use of uncoated intermittent catheters per patient, the health benefit with hydrophilic-coated intermittent catheters offsets these costs, by definition a cost-effective result. This means that hydrophilic-coated intermittent catheters are likely to be a cost-effective alternative to uncoated intermittent catheters. Their adoption across clinical practice could avoid a substantial number of infections, thereby freeing up healthcare resources in the National Health Service and reducing antibiotic use in urinary catheter users.

Keywords:

• Economic evaluation
• urinary tract infection
• catheter
• neurogenic bladder
• spinal cord

Introduction

Injuries and conditions affecting the spinal cord can lead to impaired bladder function and, therefore, urinary retention, whereby an individual is unable to completely empty urine from the bladder. Two examples of such conditions are spinal cord injuries (SCI) and multiple sclerosis (MS), which are estimated to affect approximately 50,000 and 130,000 people in the UK, respectively^1,2. Urinary retention is a common issue for these patient groups and can have serious consequences if left untreated, such as urinary tract infections (UTIs), urinary incontinence, bladder damage, and kidney damage, which can lead to kidney failure in some individuals³.

Methods used to assist emptying of the bladder for those who are unable to void naturally are transurethral or suprapubic indwelling catheters and intermittent catheters (ICs)³. According to the EAU guidelines and the National Institute of Health^3,4, intermittent catheterization, whenever possibly aseptic, is the gold standard. “Clean” intermittent catheterization, popularized by Lapides et al.⁵, was found to be the safest method for bladder emptying, with the lowest potential for urological complications in patients with a SCI⁶. Despite this, UTIs are still a common complication seen in catheter usage, in a retrospective study with a 12-year follow-up, chronic or recurrent urinary tract infections were present in 42% of patients performing clean intermittent self-catheterization⁷.

Catheter-associated UTIs (henceforth referred to as CAUTIs) have a negative impact on health-related quality-of-life (HRQoL); they are at best a repeated inconvenience for patients, who already suffer with the more direct consequences of their injury or condition, and at worst they are life-threatening⁸. The treatment of these infections places a substantial cost burden on the health system and contributes to the global issue of antibiotic resistance⁹. For these reasons, the prevention of CAUTIs has been highlighted as a national priority in the UK¹⁰.

Intermittent catheters with hydrophilic coating are seen to reduce some of the risk factors for CAUTIs, particularly by reducing urethral trauma from repeated use of these products¹¹. Hydrophilic coatings for intermittent catheters are often based on a hydrophilic polymer called polyvinylpyrrolidone, PVP. The PVP is dissolved in a solvent and coated onto the catheters by a dip coating or a spray coating process. After a thorough drying, the PVP coating is crosslinked and bonded to the catheter surface, to avoid removal during use and in some case during storage in an aqueous swelling medium.

The final coating properties such as the friction force or coefficient of friction (COF) for hydrophilic coatings are extremely low as the crosslinked PVP gel contains a certain or a sufficient amount of water, giving the coating a long dry-out time, which is very important for the user as a sufficient time for handling is needed before the coating starts to dry out and makes the insertion less smooth.

Furthermore, the catheter coating stays hydrated and lubricious during use (inserted in the urethra) as the aqueous swelling medium may contain an osmolality increasing compound (NaCl) making the catheter easy to move after emptying the bladder. The withdrawal frictional force is therefore lower than that of a conventional uncoated intermittent catheter (UIC)¹². Furthermore, hydrophilic-coated intermittent catheters (HCICs) may lower the risk of long-term urethral complications that could otherwise be exacerbated by repeatedly inserting a UIC. In the UK, both HCICs and UICs are indicated for single use only¹³.

In order to optimize the care pathway for IC users, decision-makers require information on the cost-effectiveness of various treatment options. Cost-effectiveness analyses involve an assessment of an intervention’s health benefits against its costs, when compared with relevant alternatives. HCICs are more expensive products than UICs, and so a cost-effectiveness analysis can conclude whether they are likely to be good value for money for healthcare providers, such as the National Health Service (NHS) in the UK. However, existing cost-effectiveness evaluations of HCICs have been inconsistent in their findings, or their conclusions have been invalidated by the inclusion of inappropriate comparators, such as multiple use catheters (off-label use, not recommended by NICE)^14–16.

The purpose of this evaluation was to investigate the cost-effectiveness of single-use HCICs when compared with single-use UICs in a population of people with SCI or MS receiving healthcare in the UK.

Methods

No ethical approval was required for this analysis.

Description of the economic model

A model was developed in Microsoft Excel to estimate the cost-effectiveness of HCICs. Health economic models allow evidence to be synthesized from a variety of sources and extrapolated over an appropriate time horizon. They also provide a framework for testing various parameter scenarios and their influence on the cost-effectiveness of an intervention, to provide a more holistic answer to a decision problem.

The cohort modelled were IC users suffering from either SCI or MS; a 50:50 split was assumed in the base case. Single-use HCICs were compared with single-use UICs. A UK perspective was taken, focusing on the NHS and personal social services (PSS) as is standard for such evaluations. Therefore, the model adopted a cost-effectiveness threshold of £20,000 per quality-adjusted life year (QALY) gained, and applied a discount rate of 3.5% to both costs and QALYs, in line with the NICE methods and process manual¹⁷.

The model used a Markov state-transition structure to predict the outcomes of a hypothetical cohort. Over time, the cohort moved from one health state (Alive) to another (Dead) based on the probability of mortality, which varied as the cohort aged. The modelling of CAUTIs, the main clinical driver of the model, was achieved by applying incidence rates to the living proportion of the cohort with a treatment effect applied for HCICs. This model structure is illustrated in Figure 1.

Figure 1. Model structure.

CAUTIs occurring in the model were stratified into six categories, according to degree of severity. The categories were based upon clinical guidance and were intended to capture any meaningful differences in resource use or HRQoL outcomes. There was no impact of catheter type on the breakdown of CAUTIs; that is, one type of catheter was not assumed to cause more severe infections than another.

The model utilized annual cycles, meaning that movements between health states (Alive or Dead) only occurred at the end of each year. Half-cycle correction was applied to account for the fact that, on average, deaths would occur at the midpoint of each cycle.

For each cycle, mean costs and HRQoL utilities were assigned to the living cohort. The proportion of the cohort in the Dead state did not accrue costs and QALYs. The model utilized a lifetime horizon (meaning that the cohort were followed until death), to ensure that all differences in costs and benefits were captured in the evaluation. Total costs and QALYs were calculated for HCIC users and UIC users, and these were summarized in an incremental cost-effectiveness ratio (ICER) amongst other statistics.

Key model inputs

The characteristics of the modelled cohort were averages of the demographics for SCI and MS populations^18,19. These characteristics were used to ensure that mortality and HRQoL estimates in the model were age- and gender-appropriate. Background mortality was sourced from Office of National Statistics life tables for England and Wales²⁰. An elevated risk of mortality was applied for the SCI/MS population using standardized mortality ratios, which were applied multiplicatively to the general population estimates^21,22. In addition, the model captured the risk of death from infection, via the number of CAUTIs predicted to be fatal. See Table 1 for general model inputs.

Table 1. General model inputs.

Parameter	Base case value	Source
Starting age	47	Average across the two indications: SCI-specific demographics^18. MS-specific demographics^19.
Proportion male	55%
Baseline rate of infections (in uncoated catheter users)	3.27	Weighted average of rates from four studies: [n = 114, annual rate = 5.76]^23, [n = 23, annual rate = 1.68]^24, [n = 61, annual rate = 4.56]^25, [n = 502, annual rate = 2.62]^26.
Relative risk of infection (treatment effect of hydrophilic-coated catheters)	0.84	^27
Standardized mortality ratio (SMR) for SCI	2.07	^21
Standardized mortality ratio (SMR) for MS	2.61	^22
Average cost per hydrophilic-coated catheter (male)	£1.78	^28
Average cost per hydrophilic-coated catheter (female)	£1.66	^28
Average cost per uncoated catheter	£1.30	^14
Cost per sachet of lubricant	£0.16	Optilube sterile lubricating jelly^28
Number of catheters used per day (both uncoated and hydrophilic-coated)	5	^4
Utility decrement for SCI catheter users	0.079	^14
Utility decrement for MS catheter users	0.327	^29
Utility decrement for uncoated catheter users (as opposed to hydrophilic)	0.017	^30

The baseline annual rate of CAUTIs was calculated as a weighted average of estimates from four randomized controlled trials^23–26, as this was considered a more robust estimate than any one of the individual results. This rate was used to predict the number of CAUTIs that occurred per cycle in UIC users. A relative rate of infection – sourced from a network meta-analysis²⁷ – captured the expected reduction in number of CAUTIs when HCICs were used as an alternative, which was applied multiplicatively to the baseline rate to give the annual rate of CAUTIs in HCIC users. Proportions informally elicited from the clinical co-authors were assigned to estimate the outcome severity of these CAUTIs, with costs and HRQoL impacts assigned accordingly. See Table 2 for infection-related inputs.

Table 2. Inputs predicting infection outcomes (independent of catheter type).

Parameter	Base case value	Source
Breakdown of UTIs
Simple infection treated in primary care	50%	Clinical opinion
Prolonged infection treated in primary care	25%
Complex infection treated in day case secondary care	3%
Complex infection treated in day case secondary care	17%
Infection with severe complications, not leading to death	4.25%
Infection leading to death	0.75%
Cost of treating each event type
Simple infection treated in primary care	£70	See Supplementary Material for further details.
Prolonged infection treated in primary care	£113
Complex infection treated in day case secondary care	£262	Weighted average of DC (day case) codes “kidney or urinary tract infections” (LA04L to LA04S) in NHS reference costs 2019/20^31.
Complex infection treated in day case secondary care	£2,879	Weighted average of NEL (non-elective inpatient) codes “kidney or urinary tract infections” (LA04H to LA04S) in NHS reference costs 2019/20^31.
Infection with severe complications, not leading to death	£5,924	^32
Infection leading to death	£10,426	Assumption: 1.76* the cost of an infection with severe complications, not leading to death.
Utility decrement for each event type
Simple infection treated in primary care	0.10	Assumption, validated by clinical opinion.
Prolonged infection treated in primary care	0.10
Complex infection treated in day case secondary care	0.15
Complex infection treated in day case secondary care	0.15
Infection with severe complications, not leading to death	0.17	^33
Infection leading to death	0.17	Assumption, validated by clinical opinion.
Duration of decrement for each event type (days)
Simple infection treated in primary care	7	Assumption, validated by clinical opinion.
Prolonged infection treated in primary care	14
Complex infection treated in day case secondary care	14
Complex infection treated in day case secondary care	28
Infection with severe complications, not leading to death	37	^27
Infection leading to death	65	^27

EQ-5D utilities were used to estimate health-related quality-of-life (HRQoL). Baseline utility for the general population was based on UK population norms³⁴. Decrements were applied for the SCI and MS catheterizing populations to capture the associated fall in utility, calculated by subtracting population-specific utilities from population norms for the general population. For people with an SCI who use catheters, a utility of 0.831 was used, sourced from a previous evaluation in this population¹⁴. For people with MS, a utility of 0.518 was sourced for somebody with an expanded disability status scale (EDSS) score of 5, as this was believed to be averagely representative of an MS patient with bladder impairment²⁹. It was assumed that this accounts for catheter usage and so no further utility decrement was applied for catheterization. From these population-specific utilities came utility decrements of 0.08 and 0.33, respectively. An additional utility decrement of 0.017, approximately equivalent to a 2% reduction in HRQoL, was applied to UIC users to account for HRQoL differences beyond those attributable to UTIs, such as improved comfort and convenience³⁰.

Each CAUTI event was associated with a utility decrement and duration of decrement, which were multiplied to provide an estimate for the QALY impact of each event. The durations should reflect the length of recovery for each event. To reflect the HRQoL impact of infections with severe complications, a utility decrement for sepsis was sourced from a NICE submission³³. The durations of these severe infections are not evidenced for a UK setting so were sourced from Italian DRG data, and were considered by a UK clinical expert (and co-author) to be plausible in an NHS setting²⁷. For the less severe infections, utility decrements and the corresponding durations were modeler assumptions validated by clinical opinion.

Unit costs were combined with the frequency of catheter use to produce an average cost of catheterization for each treatment arm. The NHS Drug Tariff (November 2021) was used to source unit costs for commonly used HCIC catheter brands, and a simple average product cost was calculated for each sex²⁸. The average unit cost of a UIC, also derived from the NHS drug tariff, was sourced from an existing evaluation, and was inflated to match the cost year of this evaluation (2020/2021). One sachet of lubricant was costed for every UIC used. Five catheters were costed for per day, based on the EAUN recommendation of four to six catheters per day⁴.

For infections treated in primary care, a micro-costing approach was taken whereby the costs of each element of resource use (antibiotics, GP time, testing) were sourced and used to calculate an overall cost for each type of infection. NICE guidance (2018) was used to determine appropriate antibiotic dosage for the treatment of a CAUTI³⁵. Antibiotics were costed using the drugs and pharmaceutical electronic market information tool (eMIT) or the British National Formulary (BNF)^36,37.

An equal split in use of Nitrofurantoin, Trimethoprim and Amoxicillin was assumed for first-line antibiotic treatment in line with NICE guidance³⁵; the same approach was taken with Pivmecillinam, Cefalexin, Co-amoxiclav, and Ciprofloxacin for second-line treatment. Additional healthcare-related costs associated with CAUTI treatment in primary care were identified, and clinical input was sought to obtain the corresponding resource use frequencies. These frequencies were multiplied with the corresponding unit costs and summed to produce a total cost of treating each type of infection. The inputs used in this micro-costing approach are provided in the Supplementary material.

For infections treated in secondary care, composite costs of treatment (incorporating all elements of resource use) were sourced for use in the model. The NHS reference costs were used where these could be aligned with the description of outcomes in the model³¹. The upper estimate of a literature-sourced cost of treating a catheter-associated blood stream infection (CABSI) was assumed reflective of the cost of treating an infection with severe complications (non-fatal), and was inflated to match the cost year of this evaluation (2020/2021)³². For the cost of a fatal infection, an assumption was made to maintain proportionality with the HRQoL outcomes. The duration of an infection of this severity in the model is 1.76* the duration of an infection with severe complications, not leading to death. To maintain proportionality in the model, the cost applied is assumed to be 1.76* the cost of an infection with severe complications, not leading to death.

Sensitivity analyses

Deterministic sensitivity analysis (DSA) was performed to account for parameter uncertainty, with the outcomes presented in the form of a Tornado plot. Univariate DSA demonstrates the impact of changes in individual parameters, whilst all others remain fixed, giving an indication as to which parameters have the greatest impact on the model results.

Probabilistic sensitivity analysis (PSA) was also performed to account for multivariate uncertainty in the model. In PSA, parameter values are generated from a pre-specified distribution based on type of variable, and the results are summarized to provide an estimate of the probability of cost-effectiveness. The PSA in this evaluation was run with 10,000 iterations. A cost-effectiveness plane and a cost-effectiveness acceptability curve were generated.

Results

The results of the base case evaluation, as estimated by the mean of all probabilistic iterations, are summarized in Table 3. The results reflect that HCICs are £3,183 more expensive than UICs overall but are associated with 0.55 more QALYs over a patient’s lifetime (both results have been discounted). The average survival is greater for the HCIC arm, a result that is driven by a reduction in infection-related fatalities.

Table 3. Results of the evaluation.

Cost-effectiveness results per patient (over a lifetime) – means and 95% credible intervals
	Hydrophilic		Uncoated		Incremental
Cost per patient (discounted)	£75,322 (£60,440–£92,198)		£72,139 (£57,128–£89,566)		£3,183 (−£8,293–£14,665)
QALYs per patient (discounted)	8.05 (6.48–9.43)		7.50 (5.88–8.95)		0.55 (0.32–0.82)
Life years per patient (undiscounted)	20.07 (15.52–23.92)		19.10 (14.31–23.35)		0.97 (0.30–1.78)
Incremental cost effectiveness ratio (ICER)					£5,755 (−£15,521–£32,644)
Net monetary benefit (NMB), at a threshold of £20,000 per QALY gained					£7,878 (−£5,210–£20,616)
Probability of cost-effectiveness, at a threshold of £20,000 per QALY gained					89%
Predicted event counts per patient (over a lifetime)
	Hydrophilic	Uncoated		Incremental (events avoided)		NNT*
Simple infections, treated in primary care	27.3	30.9		3.5		1
Prolonged infections, treated in primary care	13.7	15.4		1.8		1
Complex infections requiring day case in hospital	1.6	1.9		0.2		5
Complex infections requiring inpatient stay in hospital	9.3	10.5		1.2		1
Infections with severe complications, not leading to death	2.3	2.6		0.3		4
Infections leading to death	0.41	0.46		0.05		19

*The number needed to treat (NNT), represents the number of people who would need to use HCICs rather than UICs (for their remaining lifetime) in order to avoid one such event.

The estimated ICER is £5,755 per QALY gained. This result is comfortably below the cost-effectiveness threshold in the UK of £20,000 per additional QALY gained, meaning the model finds HCICs to be cost-effective when compared with UICs in this population.

Net monetary benefit (NMB) represents the net benefit of HCICs compared with UICs, when all benefits are converted into monetary units. It was calculated on the basis of a willingness-to-pay of £20,000 per additional QALY gained. The NMB is £7,878, its positive value reflecting the favorable cost-effectiveness result.

HCICs were cost-effective in 89% of simulated cases. The graphical PSA outputs (cost-effectiveness plane and cost-effectiveness acceptability curve, both in Figure 2) illustrate and summarize the results of the 10,000 iterations.

Figure 2. Top panel ∼ Cost-effectiveness plane. Bottom panel ∼ Cost-effectiveness acceptability curve.

Also presented in Table 3 is the number of CAUTI events predicted over a patient’s catheter-using lifetime for each treatment arm in the model, and the differences between them. The number needed to treat (NNT) is also presented for each event type; this indicates the number of people who would need to use HCICs rather than UICs (for their remaining lifetime) in order to avoid one such event. In total, seven CAUTIs are avoided; of these, 1.8 would require secondary care (either day-case or inpatient). Moreover, just one patient would need to be prescribed HCICs rather than UICs in order to prevent a hospitalization.

The results of the deterministic sensitivity analyses are presented in Figure 3. Key findings were that even using the most conservative estimate of treatment effect (informed by the 95% confidence interval reported by Rognoni et al.¹⁶), HCICs are still estimated to be cost-effective. Similarly, when the HRQoL benefit for patient preference is removed entirely (set to zero), HCICs are still estimated to be cost-effective. The cost of catheterization was the most influential parameter and the only one with potential to change the conclusion of the evaluation; if the cost of hydrophilic-coated catheters were to increase by 25% then these costs would no longer be offset and deemed good value for money, under the aforementioned cost-effectiveness threshold.

Figure 3. Tornado diagram.

Discussion

The evaluation found HCICs to be cost-effective when compared to UICs in the modelled population. HCICs are not a cost-saving intervention, despite the events avoided, but the additional costs associated with their use are countered by additional QALYs and would be considered good value for money by NICE¹⁷. Whilst there is uncertainty around model parameters, the above conclusion holds in 89% of simulated cases.

The positive cost-effectiveness result of our evaluation was driven by two key benefits of HCICs. Firstly, there is an established reduction in risk of CAUTIs for people who use HCICs rather than UICs. The magnitude of this reduction was sourced from a network meta-analysis, the gold standard of evidence²⁷. This translates to a HRQoL benefit for patients as well as cost savings for the healthcare system. Another consequence of the reduction in CAUTIs is that patients are expected to live longer, as their risk of infection-related mortality is reduced.

The other key benefit captured in this evaluation was the relative improvement in HRQoL directly associated with use of HCICs, compared with UICs. This benefit is not attributable to a reduction in infections; it instead reflects the perceived value of fewer steps, convenience, and improved comfort of the catheterizing procedure with an HCIC. The inclusion of this patient satisfaction benefit is aligned with recommendations from the International Consultation on Incontinence as well as existing evaluations in this area^14,38,39.

In spite of the evidence supporting the use of these parameters, a conservative standpoint was taken in the evaluation and all reasonable efforts were taken to avoid overstating their impact. The conclusions of the analysis, however, were robust to the most conservative values assessed for these parameters, as can be seen from the results of the DSA.

In fact, the most uncertain parameters in the model – those informally elicited from clinical opinion – were not considered key drivers of the incremental results and thus this parameter uncertainty was relatively unproblematic for determining cost-effectiveness. Conversely, the cost parameters identified as being highly influential (demonstrated by their position at the top of the Tornado diagram, Figure 3) are known with a reasonable amount of certainty, as the base case unit costs were sourced from national cost databases. Micro-costing was used to compute catheterization costs, and given the resultant importance to the model results, these costs were varied by 25%, as opposed to 20%, to incorporate potential overhead costs for HCICs and UICs. It is unlikely that one type of catheterization would be subject to a 25% cost increase without a corresponding cost increase in the other type of catheter. As a result, the cost-effectiveness results can be considered robust to individual cases of parameter uncertainty.

Notably, the conclusion of our evaluation differs to that of an existing evaluation by Bermingham et al.¹⁴, who found that UICs were the most cost-effective of the options they assessed. However, in their evaluation UICs were reused, which is off-label use. The NICE guideline development group, who were basing their recommendations upon Bermingham et al.’s research, determined that there was insufficient evidence to inform a recommendation for UICs as a reusable product. In scenario analysis, Bermingham et al. found that UICs ceased to be cost-effective when more than two catheters per day are used. Their evaluation also included gel reservoir catheters as an option, which were found to be slightly more effective than HCICs with a small additional cost and an ICER of approximately £3,000 per QALY gained. The guideline group ultimately determined that patients should be given a choice between hydrophilic and gel-reservoir catheters⁴⁰. Our analysis did not include a comparison against gel-reservoir catheters for this reason; it is accepted that both options should be available for patients.

Another more recent evaluation by Clark et al. produced results very similar to ours, despite differences in model structure and parameterization¹⁵. Whilst their evaluation did not factor in patient satisfaction, it included a larger reduction in CAUTIs and so the overall QALY benefit was close to the one predicted by our evaluation. A further evaluation by Rognoni and Tarricone¹⁶ also estimated HCICs to be cost-effective, despite, again, excluding the patient preference benefit.

Strengths and limitations of the analysis

Informal expert elicitation was utilized at several key points during model development. Firstly, the proposed model structure was presented to the clinical co-authors to ascertain how well it reflected the clinical pathway. All economic models are simplifications of reality, but it was deemed that this de novo model was appropriate in its assumptions and that its structure allowed the main differences in cost and benefits to be captured.

In addition to validation of the model structure, the clinical co-authors also participated in validating the key model inputs. As with structural limitations, some parameter uncertainty is inevitable in economic models. Where evidence was scarce and modeler assumptions were required, these assumptions were, wherever possible, confirmed to be clinically plausible.

The main limitations of this evaluation relate to the simplification of the treatment pathway that was necessary in developing a useful tool for decision-making. For instance, the issue of microbial resistance has been well-documented in the literature, and the costs associated with an antibiotic-resistant infection have not been captured explicitly in this evaluation⁴¹. The results are, therefore, possibly understating the true benefit of the reduction in frequency of CAUTIs attributable to hydrophilic-coated catheters over uncoated catheters.

Moreover, long-term effects of UTIs, such as urinary incontinence, bladder damage, kidney damage, and kidney failure were not included within the analysis, which is a source of conservative bias.

Further, many CAUTIs may be considered recurrent, and it is likely that recurrent infections are associated with a higher cost (for example due to referrals to specialists, additional monitoring) as well as a greater impact on HRQoL. This, too, was not captured in the evaluation, and is another potential source of conservative bias in the results.

The progression of the underlying condition of the cohort was not modelled; it was assumed that the effect on HRQoL and mortality associated with MS, or an SCI was constant over time. This was a necessary simplifying assumption and should not substantially bias the incremental results.

In terms of the generalizability of the results of this evaluation, we acknowledge that the modelled cohort may not entirely reflect the population of interest. For instance, an assumption was made that the cohort should be constructed of an equal split of individuals with SCI and with MS. However, the cost-effectiveness results are robust to changes in this population breakdown; for instance, a cohort of 100% SCI individuals produces very similar results.

Further, the starting age of the MS portion of the cohort reflects the mean age of MS catheter users, and not the mean age of catheter initiation in these patients. This potential overestimation in the starting age effectively reduces the period over which hydrophilic-coated catheter users can accrue incremental benefits compared with uncoated catheter users and is therefore another source of conservative bias.

Overall, the limitations are likely to have led to an underestimation of the cost-effectiveness of hydrophilic-coated catheters, which strengthens the overall conclusion of the analysis. Nonetheless, these potential sources of bias should be considered when interpreting the results of this evaluation.

Conclusions

In summary, HCICs are highly likely to be a cost-effective alternative to UICs, a result that is driven by substantial reductions in the number of CAUTIs. Their adoption across clinical practice could avoid a substantial number of infections, thereby freeing up healthcare resource in the NHS and reducing antibiotic use in users of urinary catheters.

Transparency

Declaration of funding

Coloplast sponsored this economic analysis and contributed to the development and review of this article.

Declaration of financial/other relationships

Hannah Baker, Brooke Avey and Stuart Mealing are employees of York Health Economics Consortium, who were commissioned by Coloplast to provide consultancy, develop the economic model, and draft the manuscript. Line Overbeck Rethmeier and Marie Lynge Buchter are employees of Coloplast. Marcio Averbeck is a paid advisor to Coloplast, Medtronic, Boston Scientific, and GSK. Nikesh Thiruchelvam is a paid advisor to Coloplast. He also receives teaching fees from Boston Scientific. Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

All authors contributed to the model design, parameter elicitation and validation of results. Hannah Baker, Brooke Avey and Stuart Mealing were responsible for model construction. All authors contributed to the manuscript draft.

Acknowledgements

None.