Switching antipsychotic medication to reduce sexual dysfunction in people with psychosis: the REMEDY RCT

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 16/95/01. The contractual start date was in May 2018. The draft report began editorial review in January 2020 and was accepted for publication in May 2020. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Michael J Crawford is Director of the College Centre for Quality Improvement at the Royal College of Psychiatrists (London, UK) (from 2011 to present) and has been a member of the National Institute for Health Research (NIHR) Health Technology Assessment General Committee (2017–18). Elizabeth Hughes received personal fees for speaking at a Lundbeck (Copenhagen, Denmark)-sponsored event. Zoe Hoare has been a member of the NIHR Health Services and Delivery Research panel (2016–20). Carol Paton received personal fees as an advisory board member for Allergan Ltd (Marlow, UK). Sofia Pappa received personal fees as a speaker for Janssen Pharmaceutica (Beerse, Belgium), Sunovion (Marlborough, MA, USA) and Recordati (Milan, Italy). Thomas RE Barnes received personal fees as a speaker for Janssen and as an advisory board member for Lundbeck, Newron Pharmaceuticals (Milan, Italy) and Gedeon Richter (Budapest, Hungary).

Copyright statement

© Queen’s Printer and Controller of HMSO 2020. This work was produced by Crawford et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

2020 Queen’s Printer and Controller of HMSO

Impact of sexual dysfunction associated with antipsychotic medication

At least half of people who take antipsychotic medication for schizophrenia experience sexual dysfunction. 6–8 The mechanisms through which antipsychotic drugs cause sexual dysfunction are varied. 9,10 Many antipsychotic drugs block dopamine receptors. This can lead to increased levels of prolactin, which reduces sex drive. 11 Some antipsychotic drugs have sedative effects, which may also reduce libido. Others have direct effects on blood flow to reproductive organs and can impair sexual functioning. Changes in blood lipids and glucose metabolism associated with antipsychotic medications may also lead to peripheral vascular disease and associated impairments to sexual functioning over the longer term. 12

Sexual dysfunction can have a negative impact on a person’s well-being and quality of life. 13–15 In a study of 139 men with schizophrenia in North America, those with sexual dysfunction reported lower overall quality of life than those with no sexual dysfunction. 16 Among those with sexual partners, people experiencing sexual dysfunction reported having poorer-quality relationships and being less likely to discuss their illness with their partner. 16

Sexual dysfunction can also affect adherence to antipsychotic medication. Surveys of people with psychosis indicate that sexual dysfunction is a common reason for stopping antipsychotic medication, which increases the likelihood of relapse and readmission to hospital. 17,18 Many patients find it difficult to talk to their mental health team about sexual dysfunction. For example, in one survey 80% of women who had experienced sexual dysfunction associated with use of antipsychotic medication had not discussed this with their mental health nurse or psychiatrist. 17 This means that people may stop their medication without seeking advice about how to manage these important problems.

Assessment of sexual dysfunction among people taking antipsychotic medication

Although sexual side effects of antipsychotic drugs are those that people with psychosis are most troubled by,18 there is evidence that clinicians find it difficult to discuss them. 19 Data from audits of prescribing practice show that sexual side effects of antipsychotic drugs are those least likely to be recorded in a patient’s clinical records. 20 It is unclear why mental health professionals have paid less attention to sexual side effects of antipsychotic medications than to other adverse effects. It is possible that some find it difficult to talk to patients about sexual side effects or are not aware how common they are. By contrast, most people using mental health services consider their sexual health and functioning to be important for their quality of life16 and expect mental health professionals to enquire about sexual matters. 21

Management of sexual dysfunction associated with antipsychotic medication

When sexual dysfunction is associated with the use of antipsychotic medication, clinicians may try reducing the dose of medication they prescribe. 9 This can be an effective strategy when it is possible to do this without having a negative impact on the person’s mental health. 22

When reducing the dose of medication is not possible, an alternative approach is to switch to an antipsychotic that is associated with a lower incidence of sexual dysfunction. Open-label studies of switching a person’s antipsychotic medication to aripiprazole23 or quetiapine24 have demonstrated beneficial effects. However, to the best of our knowledge, only two small randomised trials have compared the effects of switching medication with no switch for the management of sexual dysfunction associated with antipsychotic medication. Kinon et al. 25 randomised 54 men and women who were taking antipsychotic medication for schizophrenia and who had sexual dysfunction to either a switch to olanzapine or continuation of their current drug regimen. The study25 noted improved self-reported sexual functioning among those switched to olanzapine at 4 months. Byerly et al. 26 compared switching from risperidone to quetiapine with maintenance risperidone on sexual functioning of 42 men and women with schizophrenia. Participants were followed up for only 6 weeks. No statistically significant differences were found, but a non-statistically significant trend towards improved sexual functioning was seen in those who switched to quetiapine. Neither study was sufficiently powered to assess harms associated with switching antipsychotic drugs.

Switching a person’s antipsychotic medication may increase the risk of relapse,27,28 and different side effects associated with the medication may lead to poorer health outcomes or reduced quality of life. 3 A Cochrane review29 of the management of sexual dysfunction associated with antipsychotic medication among people with schizophrenia concluded that switching to another drug may provide an effective way to manage sexual side effects. An update5 of this review in 2018 found no new trials that tested the effects of switching medication.

The REMEDY trial

Switching medication may provide an effective strategy for improving sexual functioning and quality of life for people with schizophrenia who experience sexual dysfunction associated with use of antipsychotic medication. However, uncertainty about how effective this strategy is, together with concerns about increased risk of relapse and emergence of new side effects, means that clinicians and patients do not know whether or not this is an effective approach to help people who experience these problems.

The Randomised Evaluation of Management of sExual DYsfunction (REMEDY) trial was designed to address this uncertainty. The main aim of the study was to test whether or not switching a person’s antipsychotic medication to one considered to have a relatively low propensity for sexual side effects plus brief psychoeducation and support improved patient-reported sexual dysfunction compared with brief psychoeducation and support alone.

Trial objectives

• To examine whether or not switching medication is a clinically effective and cost-effective strategy for improving sexual functioning among people with psychosis who experience sexual dysfunction associated with the antipsychotic medication they are prescribed.

• To examine whether or not switching antipsychotic medication leads to changes in mental health, side effects of medication, health-related quality of life or service utilisation for people with psychosis who report sexual side effects.

• To examine barriers to recruitment when conducting a trial of an intervention to reduce sexual dysfunction among people with psychosis.

Design

The study was a multicentre, two-arm, parallel-group, researcher-blind randomised controlled trial with an internal pilot and a parallel qualitative study.

Study setting

The study setting was secondary care NHS community mental health services in England. We planned to recruit participants from all those in contact with community mental health services in five Trusts in the north and north-east of England and west and north-west of London.

Participants

To be eligible to take part in the study, potential participants had to be aged ≥ 18 years, have a diagnosis of schizophrenia, schizophreniform disorder, schizoaffective disorder or psychosis not otherwise specified, as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,30 and have sexual dysfunction associated with their use of antipsychotic medication, for which reducing the dose they were taking was either not effective or not appropriate.

Potential participants were excluded if:

• they were acutely psychotic or had been acutely psychotic

• their clinical team judged that they had an underlying physical health condition that was responsible for their sexual dysfunction

• their current sexual problems began prior to their starting antipsychotic medication

• they were already taking part in another clinical trial

• they were unable to speak sufficient English to complete the baseline assessment

• they were currently prescribed clozapine (a medication used when people have not responded to other antipsychotic medications).

Interventions

All study participants were offered enhanced standard care comprising treatment as usual plus brief psychoeducation and support to discuss their sexual health and functioning. The content of these sessions was based on that published by NHS Choices. 31 Brief psychoeducation and support included written information on specific psychosexual issues and psychoeducation, and signposting to the NHS Choices website and their primary care team should participants want additional help. Those delivering the brief psychoeducation and support were asked to keep a record of the time taken to deliver it and the content of the session. All those who took part in the study continued to have access to mental health services. No restrictions were placed on the use of other treatments.

Treatment in the switch arm of the trial

In addition to enhanced standard care, those in the switch arm of the trial were offered a switch from their current antipsychotic medication to an equivalent dose of another antipsychotic medication that is considered to have a lower propensity to cause sexual dysfunction. Evidence from observational and experimental studies suggest that improvements in sexual dysfunction associated with antipsychotic medication may result from swapping to aripiprazole,23,32 quetiapine24,26 or olanzapine. 25 A meta-analysis33 of the occurrence of sexual dysfunction among people taking antipsychotic medication found that two of these drugs (i.e. aripiprazole and quetiapine) were among those least likely to be associated with sexual side effects and that these three medications (i.e. aripiprazole, quetiapine and olanzapine) were the least likely to be associated with arousal dysfunction.

We asked clinicians to use their clinical judgement to select which medication to switch to, based on the known side effect profiles of these medications, patient preference, previous response to antipsychotic medication and potential interactions with other drugs the patient may be taking. 34 For patients who were being prescribed long-acting injectable antipsychotic preparations, clinicians could decide to switch the participant’s medication to oral aripiprazole, quetiapine or olanzapine, or aripiprazole long-acting injection. This was because quetiapine is not available as a long-acting injection and participating Trusts had not approved the use of olanzapine long-acting injection in community settings during the recruitment phase of the trial.

Prescribers overseeing any switch in medication were asked to gradually discontinue the participant’s current antipsychotic while at the same time gradually introducing the new antipsychotic medication (i.e. cross-tapering) to reach an equivalent dose. The equivalent was based on the Prescribing Observatory for Mental Health-UK’s Antipsychotic Dosage Ready Reckoner – Version 6.1. 35 We asked prescribers to aim to complete this cross-tapered switch of antipsychotic medication over a 4-week period (Table 1).

Assessments

Blinding

All researchers who assessed study participants were masked to allocation status. Participants and clinical staff were aware of which arm of the trial people had been allocated to. To help maintain blinding researchers were not based with unblinded clinical team members. Prior to follow-up interviews participants were asked not to reveal the arm they have been allocated to. If any researcher became inadvertently unmasked, we arranged for another researcher to collect all further data.

Study logistics

The flow of participants through the study is presented in Figure 1.

FIGURE 1.

Study flow chart.

Data management

All quantitative data were entered onto a secure web-based database. Access was restricted by encrypting the files. Audio-recordings of interviews with patient and staff were stored at local Trusts on encrypted files. Recordings were professionally transcribed verbatim and quality checked for accuracy. Once the transcriptions were verified, the recordings were deleted. Hand-written notes were taken during interviews with patients who declined to participate in the trial. These notes were then typed up, the files encrypted and the hand-written notes destroyed.

Sample size

We based the sample size calculation on our primary hypothesis: for people who experience sexual dysfunction associated with use of antipsychotic medication, switching to alternative medication that is considered to have a lower propensity to cause these side effects will result in improved self-reported sexual functioning (as rated using the ASEX).

In their randomised trial of switching antipsychotics among 42 people with sexual dysfunction associated with use of antipsychotic medication, Byerly et al. 26 reported a mean ASEX score of 19.5 points at 6-weeks follow-up and a standard deviation (SD) of 5.7 points.

We based the sample size calculation on having 90% power to detect a 3-point difference in the ASEX score between study arms. Three points on this scale equates to a clear improvement in one of the five domains in the ASEX or marginal improvement in three areas. In addition, it is at the lower end of the range of differences (2.0–5.6 points) seen in previous small-scale studies of switching antipsychotic medication that have used ASEX. 26,51 Feedback from our service user advisory group was that a change of this magnitude would represent an important change in sexual functioning.

Using these data we calculated that a total sample of 172 participants (86 participants in the switch arm and 86 participants in the control arm) would be needed to have 90% power to detect a 3-point difference in total ASEX score (SD 6.0 points) at 26 weeks, using a 0.05 significance level. To allow for a 20% loss to follow-up we aimed to recruit 216 participants.

Statistical analyses

We were not able to analyse differences in outcomes between study arms because of the very small sample size and limited number of follow-up data collected. Instead, we report details of participant flow, recruitment figures, demographic data, outcome measure descriptive statistics and details of safety events in Chapter 3. No inferences should be made from these data regarding differences between treatment arms, and summary statistics should be treated with caution because of low samples, especially when split by randomised allocation arm.

Health economics analysis

Data on the use of health and social services were collected using a modified version of the Adult Service Use Schedule, which has been used extensively in research in similar populations. 52,53 Prior to recruitment starting during the set-up phase of the trial, the questionnaire was modified based on a literature review and in collaboration with clinical and service user members of the research team to ensure that it included all relevant services. Use of services was collected in interview using patient recall, which we considered to be sufficient for the 6-month follow-up.

For each service use item a relevant and suitable unit cost was identified, and these costs are listed in Table 3. All unit costs were for the financial year 2017/18. Costs for NHS hospital contacts were extracted from NHS reference costs,56 community health and social care costs were taken from the Unit Costs of Health and Social Care 201854 and the cost of antipsychotic medication was estimated using costs per dose per item. 36

Health-related quality-of-life data were collected using the EuroQol-5 Dimensions, five-level version,57 and the ReQoL. 58 The EQ-5D is the measure preferred by the National Institute for Health and Care Excellence,59 although there are concerns about its lack of sensitivity in people with psychotic symptoms. 60 To mitigate these concerns, we included the ReQoL,49 which is a self-report outcome measure for use in people experiencing poor mental health.

The final sample size was too small to conduct the planned economic analysis. Instead, we conducted descriptive statistics on service use data, cost data at 6-month follow-up by randomised arm and quality-of-life data at baseline and 6-month follow-up. No inferences should be made from these data regarding the size and direction of differences in service use and costs, and summary statistics should be considered in the context of very small sample sizes.

Parallel qualitative study

We originally set out to conduct a parallel process evaluation in keeping with Medical Research Council guidelines. 61 However, as the study progressed and we realised how challenging it was to recruit study participants, we changed the focus of the qualitative component of the study to explore these barriers.

We aimed to interview as many as possible of the potential participants who were eligible to take part in the study but who declined to do so. We also interviewed a purposive sample of NHS staff to ensure that the views of people with a range of different backgrounds and seniority were captured. We developed semistructured interview schedules in consultation with the service user panel. Interview schedules were designed to explore patient views about reasons for not taking part in the study, and staff views about factors that promoted and hindered their referring people to the study team. Interviews lasted for up to 1 hour. Face-to-face interviews with staff were recorded, professionally transcribed and quality checked prior to data analysis. Researcher notes taken during interviews with patients were immediately typed up and stored securely following interviews.

Data were analysed using thematic analysis. 62 The researcher (LT) used an interpretative approach for analysing the data and developed themes concerning contextual factors, mechanisms and associations that may be inherent in the interviewee’s descriptions but not overtly stated.

The results of an interim analysis of data from the first 14 interviews (11 interviews with staff and three interviews with patients) were presented to members of the Trial Management Committee in June 2019. Following this, changes were made to the content of the interview guides, to take account of feedback from the group, as well as the emerging themes. These revised guides were then used for the remaining interviews with patients and staff. In the final data collection and analysis phase, themes were subjected to critical scrutiny, revision and refinement as well as restructuring of the thematic hierarchy where more prominent themes were generated.

Service user involvement

Patients identified this topic as a research priority through their membership of the James Lind Alliance Schizophrenia Priority Setting Partnership. This group, which included patient and carer representatives, rated the question ‘How can sexual dysfunction due to antipsychotic drug therapy be managed?’ as one of their top 10 research priorities. 63

Patients helped us to develop plans for this study through taking part in a survey and a focus group. Nine people attended the focus group. Members expressed enthusiasm for the topic and said that it was often neglected by mental health services. Members felt that the primary outcome should be patient-rated sexual functioning, which should be assessed over months rather than weeks. They asked us to also include assessments of people’s mental health and side effects of medication. They told us that everyone who agreed to take part in the study should be offered some support for their difficulties.

A co-applicant with lived experience of using mental health services (CG) contributed to the development of the study through attending regular Trial Management Group meetings. After the study had started, we set up a service user panel, which met on three occasions during the first 18 months of the trial. Members of the panel helped us design a leaflet for patients and a study poster and made suggestions for community-based groups among which the study could be publicised. Panel members also helped generate a list of topics to be covered in qualitative interviews with patients and staff.

Later in the study another member of the group (SJ) joined researchers in publicising the study and meeting clinical teams to help raise the profile of the trial and encourage clinicians to refer people to the study. At the final meeting of the group, members commented on the quantitative and qualitative results of the study, helped revise a draft version of the lay summary and made suggestions for future research in this field.

Ethics approval and governance

The trial was approved by the West Midlands – Solihull Research Ethics Committee (reference 18/WM/0076). In accordance with the current revision of the Declaration of Helsinki (amended October 2000,64 with additional footnotes added 2002 and 2004), a participant had the right to stop trial treatment and to withdraw from the trial at any time for any reason, without prejudice to his or her future treatment. The investigator could also withdraw a participant from trial treatment at any time in the interests of the participant’s health and well-being or for administrative reasons. Trial follow-ups continued after treatment was withdrawn unless the participant withdrew consent.

All potential participants were provided with written and verbal information about the study before being asked to provide written informed consent to take part. All participants were offered a £10 voucher at 3 months and a £20 voucher following completion of the follow-up interview at 6 months. Participants were also reimbursed for any travel expenses.

Study progress was overseen by a Trial Steering Committee and a Data Monitoring and Ethics Committee.

Changes to trial design

• During the preparatory phase of the trial we encountered problems when trying to use software for generating an Operational Criteria Checklist for Psychotic Illness and Affective Illness diagnosis from data in case records. We therefore dropped plans to confirm the clinical diagnosis assigned to the patient by the referring team.

• Following a request from the Trial Steering Committee we sent a reminder to all doctors overseeing medication of participants in the switch arm of the trial. The doctors were advised about the importance of monitoring body weight, blood glucose, blood pressure and blood lipids in the period following any change in medication.

• In the initial phase of recruitment it became clear that some patients who would otherwise be eligible to take part in the study had been psychotic within the 3 months prior to being asked about taking part in the study. We therefore changed this criterion from ‘acutely psychotic within the last 3 months’ to ‘judged by the clinical team to be sufficiently stable to take part in the study’.

• We obtained ethics and other approvals to contact all patients who were sent a letter about the study by telephone. We attempted to contact potential participants 7–14 days after sending them a letter to check that they had received it and to ask if they were interested in finding out more about the study.

• Because it became apparent that we were struggling to recruit study participants, we changed the focus of the qualitative component of the study from a formal process evaluation to a qualitative exploration of barriers to recruitment.

• We originally planned to recruit patients from three Trusts in England during the pilot phase of the trial; however, we extended this to four Trusts in the final 3 months of the recruitment phase of the study.

• Two study participants were due to have both a 3- and a 6-month follow-up assessment when the decision was made to stop the study. We obtained approvals to complete all 6-month assessments with these participants at the 3-month visit to reduce the time and cost of a longer study.

Participant flow

Table 4 and Figure 2 detail patient flow through the study. Among the 61 patients who met with a researcher, 15 (25%) consented to be screened and 46 (75%) did not. Among the 46 patients who declined to be screened, 28 (61%) provided a brief comment about their reasons for not wanting to take part in the trial. Of these, 10 (36%) patients stated that their sexual dysfunction was not important enough to need to do something about it, 11 (39%) patients said that they were worried about the possibility of having to switch their medication, three (11%) patients wanted their medication to be changed and were worried about being randomised to the control arm of the trial. The three remaining patients declined to take part, one because of work commitments, one because they did not want to be asked ‘intrusive questions’ and one patient stated that they had started taking Viagra® (Pfizer Inc., New York, NY, USA) and told us that this had improved their sexual functioning.

FIGURE 2.

Diagram of participant flow: guided by Consolidated Standards of Reporting Trials (CONSORT) guidelines. ITT, intention to treat; NA, not applicable.

Of the 15 patients who consented to be screened, 11 (73%) were eligible and four (27%) were ineligible. Of the four patients who were ineligible, one did not have psychosis and should not have been screened. No screening data for this patient have been included in the remainder of this report. Of the 11 eligible patients, 10 (91%) were recruited and randomised into the study and one (9%) withdrew consent prior to baseline assessment and randomisation. Of the 10 patients randomised, four were allocated to enhanced standard care and six were randomised to enhanced standard care plus switch of medication (see Figure 2). Following randomisation, a further two patients withdrew from the study (one from the control arm and one from the switch arm).

A breakdown detailing the reasons for non-approach, non-consent, ineligibility, non-recruitment and post-randomisation loss is given in Figure 2. The most common reason for a researcher not approaching a patient was that they were found to be ineligible (n = 32, 86%). The main reasons for ineligibility prior to consent were not having a primary diagnosis of schizophrenia or related psychoses (n = 11, 34%), being prescribed clozapine (n = 7, 22%) or not having a suitable switch medication (n = 4, 13%). Three (9%) patients were excluded because they were being prescribed aripiprazole long-acting injection, three (9%) patients were judged by the clinical team to be insufficiently stable to take part in the study, two (6%) patients had an underlying physical health condition responsible for their dysfunction and one (3%) patient had not had a trial of dose reduction to see if this improved their sexual functioning.

The main reason for non-consent was the participant declining (n = 32, 70%). Following consent one participant was found not to have a primary diagnosis of psychosis and three patients did not meet the threshold score on the ASEX for sexual dysfunction.

Descriptive statistics

Patient and demographic details of the 14 study participants are detailed in Table 5. For patients who were randomised (n = 10) the data are split into randomised allocation arm. As indicated in Participant flow, 15 patients were screened; however, data for one patient, who should not have been screened, were not entered into the database, and therefore data on 14 screened patients are presented. Of these patients, four were not randomised (three were ineligible and one withdrew from the study) and 10 were randomised.

Data from the Alcohol, Smoking and Substance Involvement Screening Test revealed that six participants smoked tobacco (three in the control arm and three in the switch arm) and eight participants consumed alcohol (three in the control arm and five in the switch arm). No participant was currently using any other substance but five had a lifetime history of cannabis use, three had a lifetime history of amphetamines use, three a lifetime history of hallucinogenic drugs use, two a lifetime history of cocaine use and two had a lifetime history of opiates use.

The mean ASEX score for the 14 participants who were screened was 20.6 points (SD 4.97 points). Among the four participants who were not randomised, the mean score was 17.0 points (SD 6.88 points) and among the 10 participants who were randomised the mean score was 22.0 points (SD 3.46 points).

Table 6 describes the status of the patients screened according to whether or not they were eligible and, if eligible, whether or not they were randomised and, if randomised, whether or not they completed baseline data collection, and 3- (the primary outcome only) and 6-month follow-up data collection. Complete baseline data were collected for all 10 randomised participants, complete 3-month follow-up data were collected for eight participants and complete 6-month follow-up data were collected for seven participants.

Six-month follow-up data could not be collected for the two participants randomised in the final months of the study (see footnote a in Table 6) but instead 6-month secondary outcome data were collected at the 3-month follow-up time point. Therefore, ASEX score was available for all 10 participants at baseline, for eight participants at 3 months and for six participants at 6 months (two participants did not reach the point when their 6-month interview was due because the study closed earlier than originally planned).

Uptake of allocated treatments

Antipsychotic medication was switched for five out of the six participants in the switch arm of the trial (three to oral aripiprazole and two to aripiprazole long-acting injection) and one six participant withdrew from the study before a switch could be initiated. None of those in the control arm of the trial had a change in their antipsychotic medication during the follow-up period.

Nine study participants were offered brief psychoeducation and support and one six participant withdrew from the study before they could be offered it. Seven (78%) of these nine participants attended a session. Sessions lasted between 30 and 60 minutes (mean duration of 45 minutes). The main problems discussed at sessions were lack or loss of desire for sex (n = 4), erectile dysfunction (n = 3) and orgasmic dysfunction (n = 2). All participants who attended a session were told about the prevalence and possible causes of sexual dysfunction and given information about the NHS Choices website and other sources of help. All seven participants were offered the possibility of a follow-up session. One participant accepted this offer.

Side effects of medication

Table 11 provides the frequencies that patients scored ‘absent’, ‘mild’, ‘moderate’ or ‘severe’ to side effects at baseline and at 6-month follow-up using ANNSERS. The most common side effects reported at baseline were lethargy/lassitude, daytime sleepiness or difficulty waking, loss of energy or drive, problems with memory, problems with concentration and dry mouth.

Safety data, protocol compliance and withdrawals

Tables 12 and 13 contain data on the number of adverse events, protocol deviations and withdrawals noted in the study. There was one adverse event and one serious adverse event, both in the switch arm of the trial. Neither were thought to be related to study procedures. The serious adverse event was for hospitalisation for an unrelated physical health condition. Three withdrawals occurred, one before randomisation and two after, one of which was because of the serious adverse event. The other two withdrawals were due to patients withdrawing consent. One protocol deviation was noted, which occurred when a patient was screened who did not meet eligibility criteria because they did not have schizophrenia or a related psychosis.

Economic evaluation

Service use data were collected at follow-up for eight participants (four participants in the control arm and four participants in the switch arm of the trial). All resources used by study participants over the follow-up are summarised in Table 14 and the costs of these services are provided in Table 15.

The substantial difference in total costs over the 6-month follow-up period between randomised arms is the result of one participant, in the control arm, being an inpatient throughout this period. Mean scores on measures of health-related quality of life at baseline and follow-up are presented in Table 16.

Emerging themes from interviews with patients

The small number of qualitative interviews we conducted with patients means that we are unable to report fully developed themes. Instead, we present emerging themes. Potential participants who decided not to take part in the study did so having weighed up concerns about their sexual functioning with concerns about their current mental health and loss of control about their treatment if they took part in the trial. Because all the qualitative patient interviews were with those who declined to take part, these emerging themes are from a biased sample and reflect barriers to taking part and not the thoughts of all patients prescribed antipsychotic medication who experience sexual side effects. Three main themes emerged and these are discussed below.

Staff perspectives: barriers to talking about sexual functioning and the study

Staff highlighted the complex nature of sexual dysfunction among people with psychosis and discussed a range of reasons why it was difficult to refer people to the trial. Many reported that they had not referred anyone to the study despite being aware that recruitment was taking place. Barriers included reluctance to discuss sex in general as well as specific factors concerning sexual dysfunction among people with psychosis and involving people with psychosis in clinical trials. Eight main themes emerged and these are discussed below.

Staff perspectives: facilitators of discussions about sexual functioning and the study

Other staff perspectives

During the course of the interviews some staff made suggestions for future studies of sexual dysfunction among people with psychosis. Suggestions included staff training and studies examining add-on medications. Staff suggested that pharmacists, because they focus on side effects of medication, may find it easier to ask patients about these side effects. Others highlighted the importance of staff training and guidance to help facilitate discussions about sexual dysfunction.

Clinicians reported that it would be easier to start discussions about sexual dysfunction if they knew that switching medication provided an effective approach to helping people who experience these problems.

Some staff said that they already offered additional medication, such as sildenafil citrate (Viagra), and asked if this could be a topic for a future study. They suggested that offering additional medication could avoid some of the concerns they and their patients had about switching their antipsychotic medication:

When we have tried medications . . . with less side effects and these medications haven’t worked because [we] haven’t been able to control the psychic symptoms then we know we probably need something else, and the Viagra and other drugs maybe would be an alternative.

1ST028

Strengths and weaknesses

To the best of our knowledge, the REMEDY trial is the first trial in the UK to test the impact of an intervention aimed at reducing sexual dysfunction associated with use of antipsychotic medication. We used a broad range of strategies to publicise the study. We encouraged staff to refer potential participants to the trial by presenting it at local and regional meetings. Researchers attended clinical meetings to remind staff about the study and encouraged the use of screening tools for assessing sexual and other side effects of antipsychotic medication. A person with lived experience joined some of these meetings and talked to staff about the importance of sexual functioning for people with psychosis. We also used a broad range of methods to communicate directly with people with psychosis to let them know about the study.

The main limitation of the study was the small sample size, which meant that we were unable to test differences in study outcomes between those who were and were not offered a switch in medication. We do not know how successful we would have been had we tried to recruit participants at other services. After 9 months of failing to recruit to target at three Trusts we decided to open recruitment up in a fourth Trust that had expressed particular interest in the study. Part of the rationale for this was to check whether or not the difficulties we had recruiting study participants were specific to the places where we initially tried to recruit from. This fourth Trust had a track record of successfully recruiting to other clinical trials involving people with psychosis, and feedback from local clinicians was that they could recruit patients to the REMEDY trial. Although one participant was recruited over a 3-month period, the slow rate of recruitment at this fourth Trust and the findings from our qualitative interviews with staff suggest that the barriers to recruitment we experienced were ones that are likely to exist across secondary care mental health services in England.

Another limitation was the small number of qualitative interviews we completed with patients. It is inherently challenging to recruit people to take part in an in-depth interview when they had already indicated that they do not want to take part in a study. We attempted to overcome this limitation by gathering brief comments from all potential participants, including those who did not want to take part in an in-depth interview. We were able to collect brief comments from 28 patients who may have been eligible but declined to be interviewed. These comments supported the views of staff, that is, patients may not consider problems with sexual functioning to be a sufficiently important to risk the potential for negative effects of a switch in their medication. Others patients were anxious about not having a change in their treatment when they did want something to be done to try to improve their sexual functioning.

Comparison with previous literature

A recent systematic review5 of interventions for managing sexual dysfunction associated with antipsychotic medication identified six RCTs. The authors concluded that there was insufficient evidence to make recommendations for clinical practice based on the results of these trials. This was, in part, because previous studies have been small and underpowered. The six trials had between 10 and 50 participants, and the authors speculated that it ‘may be difficult to recruit participants with antipsychotic-related sexual dysfunction’ to clinical trials. 5 The results of the REMEDY trial provides further support for this suggestion and, to the best of our knowledge, for the first time, provide a rich source of data about possible reasons for this.

One of the main reasons for poor recruitment was that clinical staff found it difficult to ask patients about sexual functioning. A number of surveys have reported that clinical staff believe that sexual functioning is an important component of health and well-being, but they are reluctant to raise the subject with their patients. For instance, a survey of 100 mental health nurses in Sweden found that two-thirds believed that they had a responsibility to talk to patients about sexual concerns, but only 20% did so in practice. 65 Similar surveys in the UK also report that mental health nurses believe that it is important to discuss sexual functioning with patients. 66 By contrast, retrospective audits of clinical records find that patients are unlikely to be asked about sexual functioning unless they raise it themselves. 19,20 In-depth interviews with 14 mental health nursing staff members in the UK found that staff avoided talking to patients about sex because they did not think it is an important priority or believed that others were better placed to have these discussions. 67

When staff were able to talk to patients about sexual side effects of antipsychotic medication and identify potential participants, many patients were reluctant or unwilling to take part because of concerns about a possible switch in their medication. Very little research has examined the personal experiences of people prescribed antipsychotic medication. Research that has been conducted suggests that patient experiences are complex and that problems with side effects are a major concern. 68–70 Patients who feel settled on their medication may be fearful about the consequences of a change. 71 Such concerns are well founded given evidence that switching antipsychotic medication may increase the risk of relapse27,28 or cause new side effects. 3

Implications for future research

Although the costs and benefits of switching antipsychotic medication to help people with psychosis who experience sexual dysfunction associated with the use of these medications remain unclear, we are not able to recommend future studies of using this approach in the NHS because of the substantial problems we had with recruitment. Indeed, any trials of interventions to manage sexual dysfunction among people with psychosis are likely to be challenging unless there are changes within mental health services and in society at large to make it easier for staff and patients to talk about sex and sexual functioning.

Patient and staff concerns about switching antipsychotic medication were an important barrier to recruitment in this trial. Patients may be more willing to consider taking part in studies examining the use of adjunctive medication. Although adjunctive antipsychotic medications (e.g. aripiprazole) may lead to reductions in prolactin, which have the potential to improve sexual functioning,72 concerns about co-prescribing different antipsychotic drugs73 may limit the willingness of patients and staff to support such a study. There is evidence from a small randomised trial74 that phosphodiesterase inhibitors can improve sexual function among married men with schizophrenia who experience sexual dysfunction associated with antipsychotic medication. There is also evidence that phosphodiesterase inhibitors can improve sexual dysfunction among women in the general population. 75 If it were possible to overcome barriers to assessing sexual dysfunction among people with psychosis then consideration could be given to testing the acceptability, clinical effectiveness and cost-effectiveness of phosphodiesterase inhibitors.

Contributions of authors

Michael J Crawford (https://orcid.org/0000-0003-3137-5772) (Professor of Mental Health, Centre for Psychiatry, Imperial College London) is the chief investigator of the study and contributed to the design of the study, the study protocol and the preparation of this report.

Lavanya Thana (https://orcid.org/0000-0002-8343-5549) (Research Associate, Imperial College London) developed plans for the qualitative component of the study, contributed to the design of the study, led the analysis of qualitative data and contributed to the preparation of this report.

Rachel Evans (https://orcid.org/0000-0003-0960-9843) (Statistician, University of Bangor) led the development of the analysis plan for the study and the analysis of quantitative data and contributed to the preparation of this report.

Alexandra Carne (https://orcid.org/0000-0001-7526-8077) (Research Associate, Tees Esk and Wear Valleys NHS Foundation Trust) recruited and followed up study participants and contributed to the preparation of this report.

Lesley O’Connell (https://orcid.org/0000-0002-4580-0840) (Research Associate, Imperial College London) recruited and followed up study participants and contributed to the preparation of this report.

Amy Claringbold (https://orcid.org/0000-0001-5041-4621) (Senior Clinical Trials Co-ordinator, Imperial College London) co-ordinated the trial, contributed to the development of the study protocol and contributed to the preparation of this report.

Arunan Saravanamuthu (https://orcid.org/0000-0002-8924-724X) (Research Associate, Central and North West London NHS Foundation Trust) recruited and followed up study participants and contributed to the preparation of this report.

Rebecca Case (https://orcid.org/0000-0003-4080-4807) (Research Associate, Central and North West London NHS Foundation Trust) recruited and followed up study participants and contributed to the preparation of this report.

Jasna Munjiza (https://orcid.org/0000-0003-2368-9464) (Consultant Psychiatrist, Central and North West London NHS Foundation Trust; Honorary Senior Clinical Lecturer, Imperial College London) was the principal investigator and contributed to the conduct of the study and to the preparation of this report.

Sandra Jayacodi (https://orcid.org/0000-0001-5028-3465) (expert by experience) helped to publicise the study and interpret study findings and contributed to the preparation of this report.

Joseph G Reilly (https://orcid.org/0000-0002-2469-9878) (Consultant Psychiatrist, Tees Esk and Wear Valleys NHS Foundation Trust) was the principal investigator and contributed to the design and conduct of the study and to the preparation of this report.

Elizabeth Hughes (https://orcid.org/0000-0002-4480-0806) (Professor of Mental Health, University of Leeds) was the principal investigator and contributed to the design and conduct of the study and to the preparation of this report.

Zoe Hoare (https://orcid.org/0000-0003-1803-5482) (Principal Statistician, Senior Lecturer, University of Bangor) contributed to the design of the study and study protocol, to development of the analysis plan, to the analysis of quantitative data and to the preparation of this report.

Barbara Barrett (https://orcid.org/0000-0002-0270-6256) (Reader, King’s College London) developed plans for the economic evaluation, contributed to the design of the study and the study protocol and contributed to the preparation of this report.

Verity C Leeson (https://orcid.org/0000-0001-7548-8096) (Clinical Trials Manager, Imperial College London) contributed to the development of the study protocol and the preparation of this report.

Carol Paton (https://orcid.org/0000-0001-7756-1031) (Honorary Research Fellow, Imperial College London) was the lead pharmacist on the project and contributed to the design of the study, the study protocol and the preparation of this report.

Patrick Keown (https://orcid.org/0000-0003-4727-5880) (Honorary Clinical Senior Lecturer Newcastle University) was the principal investigator and contributed to the design and conduct of the study and the preparation of this report.

Sofia Pappa (https://orcid.org/0000-0002-6303-1547) (Consultant Psychiatrist, West London NHS Trust; Honorary Senior Clinical Lecturer, Imperial College London) was the principal investigator and contributed to the conduct of the study and the preparation of this report.

Charlotte Green (https://orcid.org/0000-0002-1781-3070) (expert by experience) led patient involvement in the study and helped interpret study findings and prepare this report.

Thomas RE Barnes (https://orcid.org/0000-0002-2324-656X) (Professor of Psychiatry, Imperial College London) contributed to the design of the study, the study protocol and the preparation of this report.

Data-sharing statement

Anonymised data can be obtained from the corresponding author.

Patient data

This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data is vital to improve health and care for everyone. There is huge potential to make better use of information from people’s patient records, to understand more about disease, develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone’s privacy, and it’s important that there are safeguards to make sure that it is stored and used responsibly. Everyone should be able to find out about how patient data are used. #datasaveslives You can find out more about the background to this citation here: https://understandingpatientdata.org.uk/data-citation.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care.