Chronic hepatitis B virus case-finding in UK populations born abroad in intermediate or high endemicity countries: an economic evaluation

Objectives The majority (>90%) of new or undiagnosed cases of hepatitis B virus (HBV) in the UK are among individuals born in countries with intermediate or high prevalence levels (≥2%). We evaluate the cost-effectiveness of increased HBV case-finding among UK migrant populations, based on a one-time opt out case-finding approach in a primary care setting. Design Cost-effectiveness evaluation. A decision model based on a Markov approach was built to assess the progression of HBV infection with and without treatment as a result of case-finding. The model parameters, including the cost and effects of case-finding and treatment, were estimated from the literature. All costs were expressed in 2017/2018 British Pounds (GBPs) and health outcomes as quality-adjusted life-years (QALYs). Intervention Hepatitis B virus case-finding among UK migrant populations born in countries with intermediate or high prevalence levels (≥2%) in a primary care setting compared with no intervention (background testing). Results At a 2% hepatitis B surface antigen (HBsAg) prevalence, the case-finding intervention led to a mean incremental cost-effectiveness ratio of £13 625 per QALY gained which was 87% and 98% likely of being cost-effective at willingness to pay (WTP) thresholds of £20 000 and £30 000 per additional QALY, respectively. Sensitivity analyses indicated that the intervention would remain cost-effective under a £20 000 WTP threshold as long as HBsAg prevalence among the migrant population is at least 1%. However, the results were sensitive to a number of parameters, especially the time horizon and probability of treatment uptake. Conclusions HBV case-finding using a one-time opt out approach in primary care settings is very likely to be cost-effective among UK migrant populations with HBsAg prevalence ≥1% if the WTP for an additional QALY is around £20 000.


GENERAL COMMENTS
The paper aims to evaluate the cost-effectiveness of increased HBV case-finding among UK immigrants based on a one-time opt out case-finding approach in a primary care setting using a Markov approach. Authors reported their findings in compliance with Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement. The health economic evaluation requires a lot of work to analyze their data. However, the loose structure and vague reporting statement severely undermine the interpretation about the paper. This paper should be a lot of interest to the readership of the journal if authors can address the following issues: (1) The authors should clearly explain why the cost-effectiveness evaluation of increased HBV case-finding among UK immigrant population is necessary even existing HBV screening strategies are effective.
(2) The targeted population should be well-defined. For example, what is "intermediate or high prevalence levels"? Why were Pakistani/British Pakistani people chosen as base case population?
(3) Authors should introduce the Markov model more clearly, such as who and when will enter the model? (4) One of the limitations of this study is "substantial uncertainty surrounding the costs of the intervention and its effect if this casefinding intervention were scaled-up to a national level". So, why did authors not conduct two-way sensitivity analysis for intervention cost and effects to evaluate how their estimates would change and be scaled-up to a national level?
Most important is that authors have to seek professional academic editing to present the introduction and analysis more accurately and concisely.

VERSION 1 -AUTHOR RESPONSE
Reviewer 1, comment 1: The authors should clearly explain why the cost-effectiveness evaluation of increased HBV case-finding among UK immigrant population is necessary even existing HBV screening strategies are effective.
Author reply: We thank the reviewer for this important point. Unfortunately existing HBV screening strategies among migrant populations are currently inadequate. As we state in the introduction, one UK study found that only 12% of migrants born in countries with intermediate to high endemicity had been tested for HBV. This indicates that the majority are likely to be unaware of their infection and at risk of liver disease progression. We underscore this point by expanding the introduction as below: (Introduction) Although uncertain, it is also likely that a considerable number of people with chronic HBV remain undiagnosed. For example, in one study in Bristol only 12% of migrants born in countries with endemic prevalence >2% had been tested for HBV [9]. Due to the often asymptomatic nature of chronic infection, individuals with HBV infection can often remain undiagnosed until they develop advanced liver disease [10]. it is critical, therefore, that increased case-finding among UK migrant populations is enhanced to ensure timely treatment and follow-up to prevent complications from liver disease. . We now add this information to the introduction and methods. Additionally, we note that although we base our analysis on data from a study among Pakistani individuals in London, our targeted population is individuals born in countries with intermediate or high prevalence levels (>=2%), so we evaluate the impact of this intervention in populations with a range of HBV prevalences as observed among UK migrants born from these countries. We add additional explanation in the methods as below.
( (Methods) Our study evaluates the cost-effectiveness of HBV case-finding in the U.K. for individuals born in countries with intermediate or high prevalence levels (≥2%). The base case analysis uses the results from an uncontrolled study in which Pakistani/British Pakistani people registered at general practices (GPs) in London's East End were written to and invited to 'opt out' of being tested for hepatitis B and C infection. Those who did not opt out were telephoned and asked to attend a clinic for testing [19] The intervention was designed to increase the likelihood of testing for each infection, assumed in this analysis to occur over the initial model cycle of one year. After this time, the intervention effect was assumed to be zero, with the probability of testing reverting to background levels. The comparator programme or 'no intervention' was defined as the background likelihood of testing through existing routes such as sexual health or genitourinary medicine clinics, antenatal clinics or primary care [24]. Although we base our analysis on data from a study among Pakistani/British Pakistani individuals in London, we evaluate the potential impact of this intervention in populations with a range of HBV prevalences as observed among UK migrants born in countries with intermediate or high prevalence levels (>=2%).
Reviewer 1, comment 3: Authors should introduce the Markov model more clearly, such as who and when will enter the model?
Author reply: We apologize for the confusion and have added further details in the initial description of the Markov model to better clarify that it simulates a closed cohort of UK migrants born in countries with intermediate or high prevalence levels.
(Methods) The economic evaluation was undertaken using a Markov approach, where a closed cohort of UK individuals born in countries with intermediate or high prevalence levels (≥2%) move between a set of discrete health states representing HBV infection stage.
Reviewer 1, comment 4: One of the limitations of this study is "substantial uncertainty surrounding the costs of the intervention and its effect if this case-finding intervention were scaled-up to a national level". So, why did authors not conduct two-way sensitivity analysis for intervention cost and effects to evaluate how their estimates would change and be scaled-up to a national level?
Author reply: We apologise if the results were unclear as we have conducted the analysis suggested by the reviewer (see Figure 4). In the sensitivity analysis we provide the information that the reviewer suggests-specifically, a two-way sensitivity analysis with varying cost and intervention effects, showing the threshold HBV prevalence that would ensure the intervention is cost-effective.
Reviewer 1, comment 5: Most important is that authors have to seek professional academic editing to present the introduction and analysis more accurately and concisely.
Author reply: We thank the reviewer for this suggestion and have edited the introduction and analysis throughout to improve clarity and precision. We welcome any additional specific suggestions.