Intermittent urethral catheters for chronic incomplete bladder emptying in adults: late-stage assessment

The clinical review included 18 comparative studies evaluating intermittent catheters with the additional features. Across the 18 studies, from 1997 to 2024, there were 3 randomised controlled trials (RCTs), 12 crossover RCTs, 1 cross-sectional study, 1 prospective comparative cohort study, and 1 in vitro study. The follow-up durations varied across studies from 1 day to 1 year. There was variation in who performed the intermittent catheterisation across studies. Most studies were company funded or included authors who were company employees.

The included evidence covered all additional features identified. There was a difference between the feature of focus (hypothesis tested) of the study and the features present within that study. Most studies aimed to compare coated and uncoated catheters. The features of the interventions and comparators varied across studies, and some studies compared more than 2 types of catheter. There was substantial variation in outcome definitions, measures and reporting.

The committee discussed the variations in features and outcomes reported across studies, and understood that there was a lack of consensus about how features were categorised. The committee heard that there is currently no agreement on categorisation of features for intermittent catheters.

The committee recognised that some of the evidence was outdated, and this might not reflect current practice.

The committee acknowledged the lack of evidence on the effectiveness of individual features. It agreed that a component network meta-analysis (CNMA) was an appropriate method to analyse the benefits of the components.

The clinical evidence reported on 5 key outcomes, but the CNMA was feasible for 3 outcomes and isolated the effects of 4 features. When compared with a basic uncoated catheter, there was a possible reduction in urinary tract infection (UTI) with enhanced coating or specially designed packaging. Microhole zone technology may reduce the risk of haematuria (mainly relating to microscopic haematuria), whereas specially designed catheter cases may increase the risk of haematuria. But the committee noted there was no information about how a specially designed catheter case could have this effect and questioned the validity of this result. For the other 4 features (integrated drainage bag, integrated handle or marking, insertion sleeve or grip, tip protector or introducer), there was insufficient evidence to isolate their effectiveness.

The committee noted that the number of studies included in the analysis was limited because of the variation in features and outcomes reported across studies. It also highlighted the potential benefits of microhole zone technology and specially designed packaging, but the certainties were 'very low' across all CNMA outcomes. The committee discussed the clinical importance of microscopic and macroscopic haematuria. It agreed that only macroscopic and visible haematuria should be considered clinically important.

The committee acknowledged the lack of consensus on the feature categories, heterogeneity of the data, uncertainties of the outcomes, and that some evidence was outdated. It concluded that, although there is evidence for catheters with different features, there is not enough evidence of the effectiveness of individual features separately, or how each feature contributes to the reported outcomes. So, more information is needed to estimate the effect size of individual catheter features. This should include the details of the features, and plausible relationships between the features and outcomes.

Catheters listed on the NHS Drug Tariff have multiple additional features and basic features. To estimate the average cost of individual features, a multiple regression model was used based on the price of 838 intermittent catheters. The model suggested that approximately 78% of the variation in prices could be accounted for by the presence of basic and additional features. But, based on one-way sensitivity analysis, the results of the regression model were sensitive to changes. The committee noted that not all features contributed to price variation and led to price increases. But, it understood that the cost of each feature that contributed to the price of a catheter might not be a true representation, because market intervention might change feature prices.

The EAG developed a multi-state Markov model to estimate the cost-effectiveness of individual catheter features. This was informed by an economic model used in NICE's guideline on prevention and control of healthcare-associated infections. The model used a time horizon of 2 years and was restricted to people with urinary retention. The committee considered the structure of the model to be appropriate.

The model assumed that people used the same type of catheter over the 2-year time horizon. The committee noted that this assumption did not reflect people's experience. The survey results showed that most people used intermittent catheters for more than 2 years and used more than one type of catheter. The committee understood that the model focused on features alone rather than real-world analysis, and there was very limited quality-of-life data and no long-term evidence. Sensitivity analysis showed that longer time horizons (up to a lifetime) had little impact on the direction or magnitude of effect, because most of the costs attributed to the catheter and differences in quality-adjusted life-years (QALYs) were small.

The committee discussed the use of assumptions for the quality-of-life and ease-of-use parameters. The EAG explained that there were different ways of measuring ease of use, and it was challenging to attribute it to individual features. It added that there was very limited data for quality of life, and that ease of use impacted on quality of life indirectly. The committee recognised the uncertainty in these assumptions and noted it as a limitation of the model.

The committee queried the standardised mortality ratio of 5.41 used in the model, because it was used for people recovering from spinal cord injury. But, it recognised that mortality was not a key driver for the analysis because of the model's relatively short time horizon of 2 years.

Over a 2-year time horizon, the base-case model found that 4 features (integrated handle or markings, insertion sleeve or grip, microhole zone technology, and specially designed packaging) were dominant (less costly and increased QALYs). The model also found that 1 feature (specially designed catheter case) was less costly and slightly less effective, but this feature was considered cost effective at the £20,000 per QALY threshold. The cost effectiveness of enhanced coating was uncertain, and integrated drainage bag and tip protector were not considered cost effective at a £20,000 willingness-to-pay threshold. Extensive one-way sensitivity analysis did not markedly change these findings.

The committee considered the model's results and noted that all modelling was restricted to people with urinary retention. Given a lack of robust evidence and the assumptions used, the model's results were subject to parameter uncertainty. The committee concluded that the model's results needed cautious interpretation, but acknowledged that not all features correlated with price variation.

The committee considered a resource impact assessment (RIA) that calculated the financial impact of catheter features. The RIA used the results of the EAG's multiple regression model. It showed that the integrated bag feature had the highest annual cost per person, followed by tip protector and enhanced coating. Three features (insertion sleeve or grip, micro-hole zone and specially designed catheter case) had a zero cost. The committee recalled that the cost of each feature from the regression model might not be a true representation, so the costs or savings from the scenario analysis might not be realisable.