Article Text
PDF
Abstract
Background Technological advances present an opportunity for more people with, or at risk of, developing retinitis pigmentosa (RP) to be offered genetic testing. Valuation of these tests using current evaluative frameworks is problematic since benefits may be derived from diagnostic information rather than improvements in health. This pilot study aimed to explore if contingent valuation method (CVM) can be used to value the benefits of genetic testing for RP.
Methods CVM was used to elicit willingness-to-pay (WTP) values for (1) genetic counselling and (2) genetic counselling with genetic testing. Telephone and face-to-face interviews with a purposive sample of individuals with (n=25), and without (n=27), prior experience of RP were used to explore the feasibility and validity of CVM in this context.
Results Faced with a hypothetical scenario, the majority of participants stated that they would seek genetic counselling and testing in the context of RP. Between participant groups, respondents offered similar justifications for stated WTP values. Overall stated WTP was higher for genetic counselling plus testing (median=£524.00) compared with counselling alone (median=£224.50). Between-group differences in stated WTP were statistically significant; participants with prior knowledge of the condition were willing to pay more for genetic ophthalmology services.
Conclusions Participants were able to attach a monetary value to the perceived potential benefit that genetic testing offered regardless of prior experience of the condition. This exploratory work represents an important step towards evaluating these services using formal cost–benefit analysis.
- Genetics
- Diagnostic tests/Investigation
- Retina
Statistics from Altmetric.com
Introduction
Retinitis pigmentosa (RP) is the collective term for a range of inherited eye conditions with an estimated UK prevalence of one in 3500.1 The common feature of these inherited disorders is an abnormality of retinal photoreceptors resulting in progressive loss of vision. Currently, there are no symptomatic or curative treatments for people with RP.
Genetic ophthalmology services for people with RP include genetic counselling and testing.
Genetic tests can provide information to inform the prognosis and management of identified genetic conditions. To date, it has not been possible to offer genetic tests to a substantial number of people with, or at risk of RP, due to technological limitations. Next Generation Sequencing has opened up new possibilities for genetic testing and a substantially larger proportion of individuals can now be offered tests.2
Increasing pressure on healthcare budgets mean commissioners require robust information on the costs and benefits of potential service developments to inform funding decisions. Methods of economic evaluation, such as cost effectiveness analysis,3 which value the benefits of new services in terms of health gain using the quality adjusted life year are used by national decision-making bodies in the UK to inform resource allocation decisions.4 The lack of curative or symptomatic treatments poses challenges for valuing the benefits of genetic testing for RP using current evaluative frameworks where the main potential benefits may not be related to gains in health status.5 Evidence suggests that there is benefit in information per se (a non-health outcome) that a diagnostic test can confer to the individual.6 Further, outcomes of genetic testing for RP also have the potential to ‘spill-over’ to an individual's current and future family members.
Cost–benefit analysis (CBA) has been suggested as a potentially useful alternative evaluative framework to value both health and non-health outcomes.7 To use CBA it is necessary to quantify the outcomes in monetary values. However, a market for healthcare does not exist in the sense necessary to capture meaningful ‘revealed’ consumer choices from the UK patient's viewpoint. Moreover, the most useful questions are generally about understanding how to develop as yet non-existent healthcare services. Therefore, stated preference methods must be used to elicit the value of benefits. The contingent valuation method (CVM) is an example of a stated preference method within CBA which is used to elicit monetary ‘willingness-to-pay’ (WTP) values for healthcare interventions where outcomes go beyond health.8 Although the CVM has been used to value healthcare services, methodological challenges remain which have—to date—precluded its use in informing national resource allocation decisions.9 For example, the validity of WTP results has been questioned and, consequently, exploratory work using qualitative methods alongside CVM is necessary to provide supportive evidence for use of the CVM and validity of findings.10 This pilot study aimed to explore whether the CVM can be used to value the benefits of genetic testing for RP.
Methods
The CVM was used to elicit monetary values for two potentially clinically relevant hypothetical scenarios (1) genetic counselling for RP and (2) genetic counselling with genetic testing for RP. In addition, a topic guide was used to inform the collection of qualitative data. Scenarios and interview topic guide are available in the online supplementary appendices 2 and 3. Ethical approval from the North West 2 Research Ethics Committee (ref: 10/H1005/48) was granted for this study.
Scenario development
The two hypothetical scenarios (available in the online supplementary appendices) were designed by the research team and ratified by an expert panel. Service attributes for genetic counselling and testing were explored such as information about risk and inheritance, opportunity for testing family members, support and decision-making. A simple scenario using one form of RP (dominant) was described using the attributes of counselling alone and the potential added benefits of diagnostic testing to find a mutation in an affected person. To make the scenarios realistic, they were framed to demonstrate the uncertain nature of genetic counselling and testing, such that testing had a 50% chance of identifying the faulty gene. These scenarios were then feasibility tested in a prepilot sample (n=5) and the wording adjusted to help participants make the best sense of the scenarios described.
Study sample
Two purposive sample frames were used for this study: (1) individuals involved with the charity RP Fighting Blindness11 including the views of affected and unaffected members of family members affected by RP, including those at risk of developing or transmitting RP (Group A) and (2) employees of the University of Manchester representing the view of members of those with no prior knowledge of the condition (Group B). The rationale behind the recruitment of these two groups lies in underlying theoretical basis of CBA; it is important to assess the potential benefits from a societal perspective because, ultimately, it is society's resources that are drawn upon to fund the services in question. Society comprises individuals with and without experience of RP and it is hypothesised that different experience levels could influence WTP responses.
Potential participants for the Group A sample were identified through a lead contact from the charity. Snowball sampling methods were used to recruit the ‘general population’ sample.
Data collection
Participants in both groups were issued scenarios in a choice of formats (Microsoft Word document, audio and Braille versions) at least 48 h prior to interview. During the interview each participant was asked to first ‘read’ scenario (1) and asked to imagine that they had been referred to an eye clinic with suspected RP. Participants were asked if they would have genetic counselling as described. If they answered yes, then the reasons for wanting genetic counselling were recorded before they were asked how much they were WTP for this service as described. WTP values were elicited using an iterative bidding technique.12 This technique involves use of a predefined algorithm where an initial monetary amount is either doubled or halved dependent on respondents’ WTP until the highest WTP amount for each person is established. An open ended question asking for the maximum each individual would be willing to pay completes the bidding game. One of four bidding game algorithms using different initial bids were randomly selected13 for each participant. This process was then repeated for scenario 2. Further details on WTP values are provided in the online supplementary appendices.
A prepiloted semistructured interview schedule (available in the online supplementary appendices) informed by previous work in this area was used to explore reasons for stated WTP to provide evidence for validity and feasibility. Additional participant-level data of interest, such as financial and demographic information, were also collected at this stage from all participants.
Data analysis
A descriptive analysis of elicited WTP values was undertaken using Stata V.12.14 A modified framework approach15 was adopted for the qualitative analysis. Following a period of familiarisation with the data, a process of data-coding and collation resulted in the identification of key themes. The final aspect of the framework approach involved an interpretation of the data, in light of identified themes, which focused on gaining an understanding and providing explanations for participant responses.
Results: sample characteristics
A summary of participant characteristics (n=52) is provided in table 1. Group A participants (n=25) were more diverse in terms of age. More than two-thirds (n=17) of Group B participants (n=27) were aged between 25 and 34 years. The sample as a whole was predominately women (65%) with similar proportions in each participant group (A=60%, B=70%).
Participant characteristics (n=52)
The overall mean household income of all participants was £43 900 per year. Group A had a greater overall mean household income (£48 100 per year) than Group B (£40 000 per year). When number of people in household was taken into consideration, income differences were slight; 52% (n=14) of Group B participants had a per person annual household income or greater than £20 000 compared with 48% (n=12) from Group A. Figure 1 shows the number of participants by group and household income band.
Household income by participant group. Access the article online to view this figure in colour.
Stated demand and WTP values
The majority (92% n=48) of participants would want genetic counselling and 86.5% (n=45) would seek testing in the hypothetical situation. A higher proportion of Group B interviewees—18.5% (5/27) compared with 8.0% (2/25) in Group A—stated that they would not want testing as described.
For counselling alone, the overall median WTP was £224.50. Median WTP value from all participants for counselling plus testing was £524.50.
Nearly 80% of those interviewed would pay at least £200 for counselling plus testing compared with just over 55% who would pay the same amount for counselling alone.
Group A's median WTP for counselling was higher than Group B's (£274.50 vs £149.50). Group A's median stated WTP was also higher for genetic counselling plus testing (£899.50 vs £275). Histograms in figure 2 illustrate ‘midpoint’ (see online supplementary appendix for details) WTP values for genetic counselling alone and that for genetic counselling plus genetic testing by participant group. Table 2 provides IQRs for overall median values and values by participant group. Mean and median values calculated with and without the inclusion of zero WTP responses are also provided in table 2. Data are positively skewed with small numbers of high WTP values. An apparent difference between groups in stated WTP values is revealed in the histograms. Fisher's exact tests in tables 3 and 4 confirm the statistical significance of the high WTP values in group A.
Median (IQR) willingness-to-pay (WTP) values in GBP (£) by participant group
Cross-tabulation of participant group and willingness-to-pay (WTP) value for genetic counselling
Cross-tabulation of participant group and willingness-to-pay (WTP) value for genetic counselling and testing
Histograms of willingness-to-pay (WTP) values for counselling and testing. GC&T, genetic counselling and testing; GC, genetic counselling. Access the article online to view this figure in colour.
Qualitative findings
Participants tended to justify their stated WTP values in terms of their current financial situations. In both groups, there were examples of stated WTP values which were limited by individuals’ budget constraints. A small number of participants from both groups linked their stated WTP to their conception of real market values for services. Similar reasons from each participant group were provided on why genetic counselling would be sought; a need to find out as much information about the condition and its implications lay behind the perceived value of counselling. Through its provision of individually-tailored information, genetic counselling was particularly valued.
Thematic analysis of perceived benefits of genetic testing
During qualitative interviews, all respondents stated that they understood the scenarios. Broadly similar perceived benefits of genetic testing were identified between participant groups and are presented under the headings of the three key themes identified: ‘value of information’, ‘acknowledgement of value to others’ and ‘influence on well-being’. Illustrative quotes for reasons behind wanting genetic testing are provided in table 5.
Reasons for wanting genetic testing
Theme 1: value of information
Participants frequently spoke about how information could reveal inheritance patterns, inform reproductive choice and facilitate reliable prognoses. Elimination of uncertainties in diagnoses was considered beneficial. This held true even when confirmation of a negative diagnosis was considered. To be equipped with definite facts could enable effective decision-making and planning. Those with prior knowledge of the condition were more likely to describe how the test could potentially improve chances for accessing future treatments or cures. Only 1 (4%) participant from Group B explicitly made the link between test results and future treatments compared with 14 people (56%) from Group A.
Theme 2: acknowledgement of value to others
No one expressed objections to genetic testing per se and everyone who answered that they did not want the test was able to justify their response for personal reasons while acknowledging that testing could be of value to others. Test results could have implications for current and future family members and this potential outcome was known and valued by interviewees in both groups. Altruistic motivations for potentially accessing genetic tests did extend beyond some respondents’ immediate families. A perception that information from test results could benefit the scientific community in its aim to develop future treatments was also to be found but mainly in Group A's responses.
Theme 3: influence on well-being
In both groups the removal of uncertainty in diagnoses was seen as a having a potentially positive influence on mental well-being. The overarching desire to be fully informed was associated with the perceived inability to get on with life until definitive facts on individuals’ condition were ascertained.
Discussion
The application of CVM proved useful in eliciting monetary amounts that individuals felt they would pay in the hypothetical situation presented. The qualitative aspect of this study facilitated an assessment of feasibility and validity of using CVM in this context.
Genetic counselling and testing for RP was considered worthwhile by participants. For the minority who would not seek a test in the situation described, justification for doing so centred on a perceived lack of value in relation to the application of results within reproductive decisions. While a slightly higher number from the group without prior knowledge of the condition expressed unwillingness to access testing, there was congruity between the two groups in the recognition of potential value to others even if no personal benefit was perceived. For this reason it may be concluded that any value for an individual in knowing that a service existed for others (eg, friends and relatives) would not be included in valuations derived from this particular type of CVM. Further studies should consider how best to account for values – including notions of ‘option’ and ‘existence’ values – which do not arise from current use.16
The overall median WTP of £224.50 suggests that genetic counselling was highly valued by participants. More than half of this sample would pay over £500 for genetic counselling and testing, indicating that a genetic ophthalmology service involving testing is preferred to one comprising counselling alone. The concurrently collected qualitative data bear this out and provide evidence for a useful context on which to build future work.
This study used a relatively small sample from which aggregated WTP values have been calculated. Further work in a larger study should more accurately reflect societal income levels. A more representative sample in relation to other variables, such as education, is also necessary. However, the number of participants from the two distinct groups has facilitated a robust qualitative analysis of a rich set of data which will ultimately inform the next phase of work in this area: designing a large scale survey of the UK population to inform a CBA.
This exploratory work provides the foundation for future studies which will seek to value the net societal benefit of genetic testing for RP. An indication of the perceived monetary value of genetic ophthalmology services for RP has been presented along with supporting evidence from the qualitative data. This work is the first and only example of the application of this methodology in this context and illustrates that it is feasible to value the benefits of genetic testing using the CVM. Previous work explored the attitudes towards predictive testing for autosomal dominant RP17 but this is the first study to explicitly aim to quantify preferences for the value of genetic testing in inherited eye conditions. Additionally, the qualitative approach employed here has the potential to allow for a deeper exploration of the topic (compared with postal survey methods) enabling new and potentially unanticipated issues to emerge.18 Furthermore, while noting that the sample is not representative of UK society, it could be argued that a broader range of views have been garnered here in comparison with the study by Mezer et al.17 Whereas in that study participants were recruited from a single hospital, half of this sample with prior experience of the condition were located across the UK and had used different genetics services across the country over a number of years. The sample of Mezer et al all had at least some experience of the condition (affected and unaffected). Participants with no prior knowledge of the condition along with those affected and related to those affected have contributed to this study. Evidence on the desirability of genetic testing has been presented. The observation that people welcome a particular service does not guarantee its commissioning. Reliable evidence on costs and benefits is required to facilitate decision-making within the NHS; genetic testing for RP has to be evaluated in a formal CBA. This study provides the necessary basis for that work to begin.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online appendix 1
- Data supplement 2 - Online appendix 2
- Data supplement 3 - Online appendix 3
Footnotes
-
Correction notice This article has been corrected since it was published Online First. The first line of the Introduction has been amended: ‘35001’ corrected to ‘3500’.
-
Acknowledgements The authors would like to thank the participants. We are grateful to Fight for Sight for their support.
-
Contributors ME collected and analysed qualitative and quantitative data and produced the first draft of the article. KP, GH, MM and GCMB designed the study. RMC collected qualitative and quantitative data. All authors contributed to data analysis and have read, contributed to and approved the final article.
-
Funding This research was funded by Fight for Sight grant number 1801 and facilitated by the Manchester Biomedical Research Centre and the Greater Manchester Comprehensive Local Research Network.
-
Competing interests None.
-
Patient consent Informed written consent was obtained from all participants prior to interviews.
-
Ethics approval North West 2 Research Ethics Committee (ref: 10/H1005/48).
-
Provenance and peer review Not commissioned; externally peer reviewed.
-
Data sharing statement Additional detail on the CVM is available in the supplementary files including a copy of the scenarios we used. Supplemental data also include the qualitative interview schedule.