The acceptability and cost of a home-based chlamydia retesting strategy: findings from the REACT randomised controlled trial

Home-based retesting and SMS reminders were found to be acceptable and home-based retesting was preferred to clinic-based retesting by the majority of home arm retesters. However 14 % of home arm retesters preferred clinic-based retesting, which may relate to confidentiality concerns in their home environment. Evaluation of costs showed that the home retesting strategy was cost saving compared to clinic-based retesting ($154 vs $169). The cost saving became more pronounced when examining the cost per infection detected due to the effectiveness of the home-based strategy in detecting more repeat infections (31 vs 12). The cost per reinfection detected via the home based strategy was $1409 per reinfection detected compared with $3133 for the clinic-based strategy.

Home-based STI screening with self-collected specimens has been shown to be feasible and acceptable in both men and women [28]. In our study, most participants preferred home-based retesting to clinic-based retesting with the main reasons relating to convenience. These findings are consistent with other studies [29, 30]. However there was a small subset of participants who preferred clinic-based retesting, particularly participants who lived with their partner, parents or alone. For those who lived with partners or parents, this may have been due to concerns about confidentiality. Another Australian study found that young people were less likely to request a home-collection kit if they lived with their parents [31]. For those living alone, concerns may have related to the size of the home test kit which was too large for an average letter box, and the inconvenience of having to collect it from the post office if no-one was home when it was delivered. In our survey, participants were also less likely to prefer home testing if they had a previous chlamydia diagnosis. This group may have developed a rapport with the clinic staff and a feeling of trust in the clinic. As has been demonstrated in other studies [21, 25, 31], a person’s social circumstances or experiences are likely to be key factors in their retesting preference, and providing options for retesting is therefore important. Another important consideration is that MSM should be offered repeat testing for other infections including syphilis and HIV. An option may be to ask MSM to return to the clinic for retesting and if they don’t return, to send them a home test kit.

There is very limited evidence about the cost of home versus clinic based retesting. In the context of the DAISY study undertaken in the USA, where young women were randomised to an intervention group to receive home testing kits for chlamydia and gonorrhoea by mail at 6, 12 and 18 months, or a control group who received a postcard at the same intervals inviting them to attend one of the participating study clinics at no cost, Smith and colleagues compared the direct and indirect costs of home and clinic based screening [32]. It was found that home collection was cost saving overall (25 USD per home test versus $111 per clinic test, including $49 in direct costs and $62 in indirect costs such as transportation and parking, child care and missed work (2005 prices) [32]. The cost savings were greater in this study as indirect costs to the patient were included, and also the purpose of the home-test component was for screening (not retesting) and did not include any additional interactions with the clinic.

In the REACT trial, taking into consideration the costs of the entire pathway, home retesting was cost saving overall at $154 per test versus $169 for clinic based retesting. The two key differences in costs between the strategies were: the cost of the retesting consultation in the clinic pathway which was slightly higher than the cost of assembling home test kits plus postage; and the laboratory fee in the clinic pathway was greater than the home strategy as a different laboratory was used. Had the laboratory costs been equal, the cost of home retesting would still have been less costly but only marginally so. The home-testing cost was likely to be an over-estimation as we conservatively included 5 min time for assembly of kits which included identifying those patients who were due to be sent a home kit and organising the kit to be mailed. If kits were pre-assembled by the laboratory and identification of patients was automated, then clinic costs would be lower. The cost of a routine program depends on how it is implemented, for example if kits were only sent to those who requested them or had not returned to the clinic, it may be less costly, but the extra complexity would need to be considered.

In the REACT trial, more infections were detected in the home arm compared to the clinic arm. The effectiveness of the home retesting strategy in detecting more repeat infections, particularly among MSM, has the potential for considerable downline cost savings to the health system through the reduction of onward transmission and reducing the risk of sequelae including HIV transmission. The cost per infection detected in the REACT trial was considerably less at $1409 for home testing versus $3133 for clinic testing, mainly due to the higher number of repeat infections detected via the home based strategy (31 vs 12). Modelling by Smith and colleagues [32] in the DAISY trial also found the cost per infection to be slightly less overall ($702 in the home arm versus $717 in the clinic arm), as fewer clinic based tests were performed (3.6 tests per person in the intervention arm versus 2.7 tests per person in the clinic arm) [32], but the difference was less than REACT as the infection rate was equal in the two arms. A strength of the study was the high survey response rate. There are some limitations to consider. We didn’t estimate the cost to the patient (indirect costs). As shown by Smith et al. in the DAISY trial, inclusion of indirect costs such as transportation/ parking, child care costs and time missed from work/ school, would probably have doubled the costs of the clinic-based retesting arm in REACT, hence cost savings for home retesting would have been greater. It is also important to consider the generalisability of these findings. Costs associated with clinic-based retesting in this study related to a sexual health clinic where patients were seen mainly by nurses. However in primary care, patients would be more likely to see a doctor with a higher salary and thus there would be even greater cost savings for the home-testing strategy in this setting. Acceptability may have been over-estimated as those who chose to respond to the survey may have been more likely to give positive responses. In addition, participants were not blinded to their study arm allocation which may have had a differential impact on the likelihood of them retesting, however there is no way to assess this. Also as the study was conducted in the sexual health clinic setting, the results may not necessarily apply to other primary care settings.

In conclusion, the study has shown that in the sexual health clinic setting, a multi-faceted intervention previously shown to be effective at increasing retesting rates, was also acceptable to patients, and cost-efficient. However some clients prefer clinic-based testing, often due to confidentiality concerns in their home environment. Both options should be provided to maximise retesting rates.