Introduction
The aim of the economic component of the Wrapped study was to assess the feasibility of the methods to collect cost and outcome data to inform the design of a future definitive RCT. Wrapped is a tailored behaviour change intervention to increase condom use among young people using web-based STI-testing services, and hence the additional resources required to deliver the intervention need to be compared with usual care. This is important because any additional benefits associated with the Wrapped intervention (e.g. in terms of reducing STI rates) must be assessed in terms of any additional costs that can be attributed to it.33
The economic costs associated with STIs are considerable, and these impact on young people, the healthcare sector, the wider public sector and society.34 STIs remain a serious public health challenge, with 401,800 new diagnoses in 2023 in England1 at a cost of around £620 million annually.35 Those most at risk are young people, GBMSM, those who identify as black and those who live in areas of higher deprivation.36 There are also important equity concerns connected with poor sexual health. The stigma which surround STIs and contraception results in specific and significant barriers to accessing services and, in contrast to other services, direct patient self-referral accounts for the overwhelming majority of attendees.37 Furthermore, in England, compared to other healthcare services, there has been a substantial reduction in sexual and reproductive healthcare budgets which means that there is a heightened pressure on costs.38
Methods
The objective of the economic analysis was to establish and assess methods for collecting the cost and outcome data for the intervention and control groups. In line with guidance from the NICE for public health interventions,39,40 we explored methods to measure costs from the perspective of the NHS, and wider public sector. The study also attempted to assess private sexual and reproductive health costs incurred by participants, for example, for the purchase of condoms or contraception.
Recruitment costs
Participants were recruited via websites offering free STI self-sampling kits to young people living in England, based in five local areas. We assumed that the costs associated with recruitment would not be required if the intervention was rolled out, as such costs would be included within local sexual health commissioning arrangements; thus the costs associated with recruitment were not included in the analysis. There is further consideration of funding and commissioning arrangements in Chapter 6.
Intervention costs
For both the intervention and the control groups, resource use data were collected prospectively by the research team. Ongoing costs associated with interventions can be considered in terms of fixed costs and variable costs, with fixed costs being those which are not affected in the short term by the number of people who use them.41 In contrast, variable costs change according to the number of people accessing the intervention; cost items such as consumables are typically variable costs.
The Wrapped intervention is complex and involves several different elements, including a website, video content and provision of condom/lube samples. The website and video content were developed in a previous study,26 and hence the costs associated with developing the website and content were not included in our analysis, as these were assumed to be ‘sunk’ costs which would not need to be reincurred if the intervention were rolled out at scale. The costs associated with the maintenance of the website element of the Wrapped intervention, and updating of video content, to ensure content continued to be appropriate and appealing to young people, were explored as part of the evaluation. For interventions which involve websites and digital content, a high proportion of the total costs are often fixed costs, for example in relation to maintaining systems and storing data.42 Such costs do not tend to vary (overall) in terms of the number of people who use them. This means that the cost of delivering the digital component of the intervention (per participant) will be determined primarily by the number of participants assumed to be accessing the intervention (the denominator), with the cost per participant assumed to be lower as the digital component is delivered at larger scale.43 Thus, reporting costs based on the number of participants included in a feasibility trial will not provide an accurate reflection of the true cost per participant if the intervention was rolled out at scale.44 It is therefore recommended that the likely take-up rate is taken into account to more accurately estimate the average costs per participant associated with an online intervention.44 We therefore analysed the costs associated with the website and video content to give a more accurate estimate of these if the intervention were rolled out.
Taking into account possible take-up rates, we assumed that the cost would be £2.00 per participant to cover the costs of maintaining the Wrapped website and associated video material, including updating content, following methods adopted previously in similar studies.45 In calculating this cost, we assumed that the Wrapped intervention could be widely used across local authorities, as it could be used alongside different models of commissioning and delivering web-based STI-testing services (see Chapter 6).
For the intervention arm, variable costs were associated with the provision of condom sample packs, condom carriers and the condom distribution service. Methods to capture the resource use and costs linked to the provision of these components were developed during this feasibility trial. This included costs relating to the assembling, packaging and postage of the sample packs, sourcing and tailoring the condom carrier, and the costs associated with the ongoing condom/lube distribution service, and any other materials used.
The Wrapped intervention is tailored to address the particular barriers reported by each user. Costs and resource use in the intervention group therefore varied according to the particular barriers to condom use reported by each participant. All participants in the intervention group were eligible to order a condom sample pack. This pack contained different makes and types of condoms and lubricant. Those participants who reported that there were issues around being able to order the type of condoms that they wanted, or with the cost of condoms, or that they felt embarrassed buying condoms, could order a maximum of 12 condoms and up to three sachets of lube each month. Those participants who stated that they did not always have a condom with them when needed were given the option to order a condom carrier.
Those randomised to the control group received usual care. This was a web page with standard information on STIs and details about how to access condoms. In line with the assumptions above, we assumed £1.00 per participant to cover the costs associated with maintaining and delivering the control website, given the scale at which this could be available across local areas.
Healthcare resource use and costs
During the fRCT, healthcare resource use by participants was collected via survey for participants in both intervention and control groups at all time points. Data were collected on whether participants had visited a sexual health clinic, their GP/nurse at a GP surgery, or a GP out of hours service, on the number of A&E visits, and any other use of health services. Unit cost estimates were applied to resource use data to estimate overall costs. The sources of unit costs were from the published literature and published reference costs.46
Public sector and private costs
Costs may fall on other sectors, apart from health, and hence we included questions to capture broader public sector resource use. In particular, we explored whether participants who reported being in education had accessed services connected with their sexual and reproductive health at school, college or university.
Other costs incurred by participants associated with their sexual and reproductive health, for example ordering and paying for condoms, STI testing, and contraception, were also captured. This included access to counselling and other services which might be provided outside of the NHS. Questions were included in the surveys at all time points to measure these costs.
Outcome data
As shown in Chapter 3, a wide range of outcome data were collected. For the health economics component, the main aim was to test the feasibility and acceptability of using a range of different outcome measures required for a future definitive economic evaluation. This included data on health-related quality of life (HRQL) collected across all time points using the EQ-5D-5L instrument, to allow a cost–utility analysis to be performed.24 Following NICE guidance, we calculated utility values by mapping the 5L descriptive system data onto the 3L value set.47 Utility values were calculated for feasibility processes only, as the trial was not powered to detect any changes in outcome measures. We also assessed the feasibility and acceptability of using the SF-12 questionnaire to measure HRQL at baseline, M3, M6 and M12.25 This is because there are limitations with using the EQ-5D in a sexual health context, particularly because many of those who are infected with a STI such as chlamydia do not experience symptoms.48 It has been suggested that the SF-12 instrument might have strengths in such contexts due to the larger descriptive system and broader coverage of HRQL.49,50
Analysis
The analysis focused on evaluating the feasibility of collecting cost and outcome data to inform the design of a definitive trial involving an economic evaluation. This involved the establishment and assessment of trial processes for obtaining cost data and the acceptability/completion of questions within the questionnaires. As this was a feasibility study, the purpose was not to produce a definitive result with respect to the cost-effectiveness of the intervention, but to conduct an exploratory analysis to design and refine the methods for an economic evaluation alongside a future definitive trial. Data on the full sample were used.
Results
Cost and outcome data were successfully collected and analysed over the full period of the fRCT. Costs were captured in relation to the delivery of the intervention and wider resource use. A range of outcome data were successfully collected and are presented below.
Intervention costs
For the intervention group, resource use linked to the provision of the Wrapped intervention was successfully collected. As detailed above, the intervention was tailored to meet the reported barriers for participants, and this meant that the resources received in the intervention group varied according to the needs of participants.
Condom sample packs
The sample pack was available to all those randomised to the intervention group. In total, 48 people ordered a sample pack. Costs associated with the products included in the sample packs and assembling costs are detailed in Table 19.
| Item | Unit cost (£a) |
|---|---|
| Condom sample pack – packaging | |
| Condom sample pack – box | 1.41 |
| Condom sample pack – leaflet printing | 0.94 |
| Condom sample pack – inlay printing | 0.34 |
| Condom sample pack – condom menu sheet printing | 0.07 |
| Total cost of packaging | 2.76 |
| Condom sample pack – contents | |
| Durex Pleasure Me | 0.53 |
| Durex Thin Feel | 0.43 |
| EXS Air Thin | 0.10 |
| EXS Extra Large | 0.08 |
| EXS Snug Fit | 0.08 |
| MyONE Perfect Fit | 1.17 |
| Pasante Cooling | 0.09 |
| Pasante Sensitive | 0.08 |
| Pasante Warming | 0.09 |
| Skyn Extra Lubricated | 0.46 |
| Skyn Intense Feel | 0.50 |
| Skyn Original | 0.29 |
| Lube – silky TLC | 0.08 |
| Lube – light/gentle | 0.08 |
| Total condom and lube costs | 4.06 |
| Other costs | |
| Postage bag | 0.13 |
| Address sticky label | 0.03 |
| Postage costs | 1.53 |
| Thank you card | 0.07 |
| Assembly costsb | 0.36 |
| Total other costs | 2.12 |
| Grand totals | |
| Total cost of one sample packc | 8.94 |
| Total orders of sample pack | 48 |
| Total cost | 429.12 |
| Cost per participant in the intervention arm | 3.73 |
Condom carrier
All those who were allocated to the intervention group, who reported that having easy access to condoms when needed was a barrier for them, could order a condom carrier. In total, 31 people ordered a carrier. The costs associated with the case are detailed in Table 20 for the full sample.
| Item: condom carrier | Unit cost (including VAT and shipping costs) (£a) |
|---|---|
| Condom carrier | 3.17 |
| Condom | 0.08 |
| Postage bag | 0.10 |
| Thank you card | 0.07 |
| Address sticky label | 0.03 |
| Postage costs | 0.88 |
| Assembly costsb | 0.36 |
| Grand totals | |
| Total cost of one carrierc | 4.69 |
| Total orders of condom carriers | 31 |
| Total cost | 145.39 |
| Cost per participant in the intervention arm | 1.26 |
Condom-ordering service
For those participants in the intervention group who reported affording condoms as a barrier to use, a condom distribution service was available. This allowed young people to select and order free condoms which were delivered by post. Over the study period, 37 participants ordered condoms, with 115 separate orders being placed. The costs associated with this condom-ordering service are shown in Table 21.
| Item | Unit cost (£a) | N | Total cost (£a) |
|---|---|---|---|
| Condoms | |||
| Durex Pleasure Me | 0.53 | 222 | 117.66 |
| Durex Thin Feel | 0.43 | 234 | 100.62 |
| EXS Air Thin | 0.1 | 60 | 6.00 |
| EXS Extra Large | 0.08 | 48 | 3.84 |
| EXS Snug Fit | 0.08 | 12 | 0.96 |
| MyONE Perfect Fit | 1.17 | 78 | 91.26 |
| Pasante Cooling | 0.09 | 30 | 2.70 |
| Pasante Sensitive | 0.08 | 60 | 4.80 |
| Pasante Warming | 0.09 | 42 | 3.78 |
| Skyn Extra Lubricated | 0.46 | 123 | 56.58 |
| Skyn Intense Feel | 0.50 | 243 | 121.50 |
| Skyn Original | 0.29 | 129 | 37.41 |
| Total cost of condoms | 547.11 | ||
| Other costs | |||
| Text messages/e-mail reminders | 0.01 | 37 | 0.37 |
| Instruction sheet | 0.60 | 115 | 69.00 |
| Thank you card | 0.07 | 115 | 8.05 |
| Postage bag | 0.10 | 115 | 11.50 |
| Address sticky label | 0.03 | 115 | 3.45 |
| Postage costs | 0.96 | 115 | 110.40 |
| Assembly costsb | 0.36 | 115 | 41.40 |
| Total other costs | 244.17 | ||
| Grand totals | |||
| Total costs for condom-ordering service (condoms and other costs)c,d | 791.28 | ||
| Number of people in the intervention arm | 115 | ||
| Cost per participant in the intervention arm | 6.88 | ||
Healthcare resource use
Costs and resource use were collected and analysed for all time points (baseline, M3, M6 and M12). In the sections below the costs collected at M12 are presented as an illustration of how the data would be used as part of an economic evaluation in a future definitive trial. As part of the study, data were collected from participants on healthcare resource use associated with sexual health. As can be seen from Table 22, the main healthcare resource use reported relates to Sexual Health clinics and GP visits.
| Resource use | N intervention group | N control group | Total |
|---|---|---|---|
| Sexual health clinic | 7 | 11 | 18 |
| GP consultation | 2 | 5 | 7 |
| GP nurse consultation | 2 | 2 | 4 |
| NHS walk-in centre | 1 | 0 | 1 |
| NHS 111 calls | 1 | 0 | 1 |
| GP out of hours | 0 | 0 | 0 |
| Pharmacy | 1 | 1 | 2 |
| A&E | 0 | 0 | 0 |
| Gynaecology | 0 | 2 | 2 |
| Other | 0 | 0 | 0 |
| Total number of participants reporting use of healthcare resources | 10 | 16 | 26 |
| Total number of participants completing this item at follow-up | 96 | 95 | 191 |
Table 23 shows the healthcare resource use reported by participants and the associated costs. It is evident that the most frequently reported resource use related to sexual health clinics and primary care.
| Resource use | Cost item | Unit cost (£a) | N wrapped intervention group | N control group | Cost–intervention (£a) | Cost–control (£a) |
|---|---|---|---|---|---|---|
| Sexual Health Clinic (face to face and telephone) | Per consultation | 136.00 | 22 | 22 | 2992.00 | 2992.00 |
| GP consultation (face to face) | Per consultation | 42.00 | 2 | 6 | 84.00 | 252.00 |
| GP consultation (telephone) | Per call | 15.80 | 4 | 7 | 63.20 | 110.60 |
| GP nurse consultation (face to face) | Per consultation | 13.00 | 3 | 1 | 39.00 | 13.00 |
| GP nurse consultation (telephone) | Per call | 8.69 | 2 | 5 | 17.38 | 43.45 |
| NHS walk in centreb | Per visit | 42.00 | 1 | 0 | 42.00 | 0 |
| NHS 111 callsc | Per call | 8.69 | 1 | 0 | 8.69 | 0 |
| Pharmacyd | Per visit | 8.83 | 2 | 2 | 17.66 | 17.66 |
| Gynaecology | Per visit | 131.00 | 0 | 4 | 0 | 524.00 |
| Total cost | 3263.93 | 3952.71 | ||||
| Number of participants completing this item | 96 | 95 | ||||
| Total cost per participant | 34.00 | 41.61 |
The study included information about STI self-sampling kit ordering and access to contraception. As is shown in Table 24, a proportion of participants reported that they had ordered STI self-sampling kits. Participants reported that most kits ordered were returned (and hence processed).
| Resource use | Unit cost (£a) | N intervention group | N control group | Total cost–intervention group (£a) | Total cost–control group(£a) |
|---|---|---|---|---|---|
| Chlamydia kit ordered | 7.27 | 29 | 10 | 210.83 | 72.7 |
| Gonorrhoea kit ordered | 7.27 | 0 | 1 | 0 | 7.27 |
| Chlamydia and gonorrhoea kit ordered | 7.27 | 20 | 5 | 145.4 | 36.35 |
| Chlamydia, gonorrhoea, HIV and syphilis kit ordered | 9.60 | 29 | 22 | 278.4 | 211.2 |
| Chlamydia kit processed | 3.00 | 28 | 10 | 84.00 | 30.00 |
| Gonorrhoea kit processed | 3.00 | 0 | 1 | 0 | 3.00 |
| Chlamydia and gonorrhoea kit processed | 3.00 | 15 | 5 | 45.00 | 15.00 |
| Chlamydia, gonorrhoea, HIV and syphilis kit processed | 6.00 | 23 | 22 | 138.00 | 132.00 |
| Total cost | 901.63 | 507.52 |
We collected data on other NHS prescriptions and services accessed by participants in relation to their sexual and reproductive health. For example, participants were asked whether they had ordered STI self-sampling kits via NHS services, or contraception. There were issues with obtaining sufficient detail on how long some products had been prescribed or accessed, and the particular nature of products. Thus, some assumptions needed to be adopted to estimate costs for services and products accessed (shown in Table 25).
| Resource use | Cost item | Unit cost (£a) | N intervention group | N control group | Total cost–intervention group (£a) | Total cost–control group (£a) |
|---|---|---|---|---|---|---|
| Contraceptive pillb | Per 3-month prescription | 4.93 | 18 | 17 | 88.74 | 83.81 |
| PEP (post-exposure prophylaxis for HIV)c | Course of treatment | 525.95 | 2 | 1 | 1051.90 | 525.95 |
| Counselling | Per course (6 weeks) | 360.00 | 6 | 4 | 2160.00 | 1440.00 |
| Medical abortiond | Per medical termination | 353.40 | 1 | 2 | 353.40 | 706.80 |
| Coil insert/removal | 42.00 | 2 | 1 | 84.00 | 42.00 | |
| Contraceptive implant/injection/patch | 42.00 | 5 | 2 | 210.00 | 84.00 | |
| Total cost | 3948.04 | 2882.56 |
As part of the fRCT, we explored the collection and analysis of costs relating to other aspects of the public sector, such as costs incurred by schools, colleges and universities. Overall, around a quarter of participants reported that they were currently studying in school, college or university (26%, 60 participants). As is shown in Table 26, a small number of participants reported accessing services at school, college or university in connection with their sexual health.
| Resource use | Cost item | Unit cost (£a) | N intervention group | N control group | Total cost–intervention group (£a) | Total cost–control group (£a) (UK £) |
|---|---|---|---|---|---|---|
| Nurse consultationb | Per consultation | 23.00 | 1 | 0 | 23.00 | 0 |
| Welfare or support worker consultation | Per consultation | 7.67 | 4 | 0 | 30.68 | 0 |
| Total | 53.68 | 0 | ||||
| Cost per participant | 0.56 | 0 |
Costs borne by participants
We attempted to capture the costs incurred by participants in relation to sexual and reproductive health, for example in terms of purchasing condoms, STI self-sampling kits, contraception etc. The costs shown in Table 27 are based on reported costs. Those in the control group reported slightly higher costs associated with sexual health.
| Resource use | N intervention group | N control group | Total cost–intervention group (£a)b | Total cost–control group (£a)b |
|---|---|---|---|---|
| Condoms | 12 | 25 | 229.00 | 415.00 |
| Lube | 29 | 31 | 326.00 | 350.50 |
| STI kits | 0 | 2 | 0 | 80.00 |
| Contraceptive pill | 4 | 4 | 105.00 | 70.00 |
| PEP | 0 | 1 | 0 | 10.00 |
| Emergency contraception | 6 | 2 | 137.50 | 26.00 |
| Counselling | 1 | 2 | 620.00 | 1040.00 |
| Total cost | 1417.50 | 1991.50 | ||
| Total number of participants who completed M12 follow-up survey | 96 | 97 | ||
| Cost per participant | 14.77 | 20.53 |
Summary of costs
Table 28 summarises key data collected during the trial. The summary is illustrative only due to the small sample size. However, the data collected demonstrate the importance of a comprehensive approach to cost collection, particularly in relation to the need to collect costs incurred by participants. As shown, participants reported that they had incurred a range of costs themselves, particularly in relation to condom ordering, contraception etc. It is worth noting that although the short-term costs below are important to understand, in the longer term these costs could be heavily outweighed by savings associated with preventing STIs and onward transmission.
| Type of resource use | Total cost – intervention (£a) | Total cost – Wrapped control (£a) |
|---|---|---|
| Intervention costs | ||
| Website | 230.00 | 115.00 |
| Sample packs | 429.12 | |
| Condom carrier | 145.39 | |
| Condom-ordering service | 791.28 | |
| Total cost | 1595.79 | 115.00 |
| Average cost per participant | 13.88 | 1.00 |
| Resource use costs | ||
| Healthcare service resource use (at 12 months) | 28.63 | 34.67 |
| Other NHS resource use – average costs per participant (reported at 12 months) | 34.63 | 25.29 |
| Resource use via schools, colleges, universities (at 12 months) | 0.47 | 0 |
| Private costs (at 12 months) | 12.43 | 17.47 |
| Average cost per participant | 76.16 | 77.43 |
Outcome data collection
As part of the fRCT, a range of outcome measures were successfully collected. This included data on health-related quality of life via the EQ-5D-5L instrument, and the SF-12 instrument. The EQ-5D-5L and SF-12 instruments were administered to inform the economic analysis, as they are preference-based measures and can be used to calculate Quality-Adjusted Life-Years (QALYs). The results are presented in Table 29. Completion rates were similar as for the other outcome measures included in the trial (see Chapter 3, Results, RO 8). As the trial was not powered to detect any changes in outcome measures over time or differences between intervention groups, the aim of this component was to explore the feasibility of collecting the health-related quality-of-life outcome measures.
| Outcome measure | Wrapped intervention Na | Wrapped intervention – Mean value (SD) | Control Na | Control mean value (SD) |
|---|---|---|---|---|
| EQ-5D-5L – Baselineb | 115 | 0.784 (0.2) | 115 | 0.768 (0.209) |
| EQ-5D-5L – Follow-up at 12 monthsb | 95 | 0.779 (0.227) | 95 | 0.758 (0.237) |
| SF-12 – Baseline (SF-6D)c | 113 | 0.681 (0.117) | 115 | 0.678 (0.114) |
| SF-12 – Follow-up (SF-6D)c | 95 | 0.691 (0.128) | 95 | 0.678 (0.109) |
Summary
This element of the study focused on the development and evaluation of processes to capture data on the costs and outcomes that would be necessary to inform the design of a future definitive trial involving an economic evaluation. Processes were successfully established to measure the costs associated with delivering both the intervention and the control. Generally, a range of comprehensive cost data were collected, including the private costs borne by the participants and resource use in other sectors such as education. The findings have implications for the design of a future definitive trial as a broad range of resource use and costs were reported, including private costs (such as for condoms, contraception and testing), which would need to be fully considered. Some difficulties were found with capturing detail on some types of resource use, such as in relation to contraception and other products. Any future definitive trial should include a cost-consequences analysis and ensure that sufficient detail is captured on resource use.
Within the trial, relevant outcome data were collected successfully, including for EQ-5D-5L and SF-12. For a future definitive trial, it would therefore be possible to consider a range of types of economic evaluation. This would include a cost–utility analysis as recommended by NICE,39,40 and in addition it would be possible to consider the inclusion of a cost-effectiveness analysis involving the main clinical outcomes included in the study. There might also need to be consideration of further outcomes that might be relevant in this context, for example around general well-being.