This is an exploratory, multicentre, individually randomised controlled trial. Figure summarises the study design and the flow of patients through the trial.
This trial involves 2 active interventions PST and ‘referral to the GP’ and a ‘waiting list control’.
The PST intervention evaluated in this study is a behavioural therapy that has been specifically designed for people with a visual impairment [8
]. This intervention involves a trained psychological therapist working with the patient in their own home or at one of the research centres. Over a 6–8
week period, patients are taught a 7-step method for approaching and solving their problems. The steps are: (1) defining the problem; (2) establishing realistic goals; (3) brainstorming solutions; (4) implementing decision making guidelines (Pros versus Cons); (5) choosing a preferred solution; (6) implementing the solution; and (7) evaluating the outcome. The PST intervention will also include additional self-help materials for patients e.g. materials concerning an explanation of depression, a discussion on the importance of treating depression, and a description of the various treatment options as well as vision related sign posting materials. This type of behavioural therapy is an individualised guided self help approach consistent with NICE guidelines [15
]. Optometrists providing the low vision assessment will share the patient’s ‘treatment plan’ with the therapist (via a brief report) so that the therapist can help the patient implement the optometrist’s recommendations via the PST framework if this is what the patient wishes. To ensure the fidelity of this intervention the PST therapists will be trained by Dr Mark Hegel the Clinical Psychologist who originally designed the PST intervention used in 2007 by Rovner, Casten et al, [8
]. With the participants consent, each PST intervention will be taped and one third of randomly selected tape recordings will be reviewed to ensure the fidelity of the intervention.
The second intervention is ‘referral to the GP’. A carefully crafted referral letter will inform the GP that their visually impaired patient has screened positive for a possible depressive disorder and invites them to offer the patient an assessment and treatment as per NICE guidelines for the management of depression. This is a pragmatic intervention and we acknowledge that GPs may or may not adhere to NICE guidelines i.e. we propose to evaluate the effectiveness of the referral, not ‘best practice’ as described by NICE. The fidelity of this intervention will be assessed by recording the number of correctly addressed referral letters sent.
The final arm of the study is a ‘waiting list control’. Patients in this arm of the trial will be referred to their GP using the referral letter described above after the 6
month outcome measure. A ‘waiting list control’ provides people taking part in the study with a better level of care than ‘treatment as usual’ because depressive symptoms are currently ignored and there is no referral. The fidelity of this intervention will be assessed by asking people at 6
months about any treatments received during the time of the study.
In addition to the interventions listed above, all participants in the study will have a follow up vision rehabilitation appointment 6
weeks after the initial appointment. Follow up appointments of this type are typical for people who appear depressed or are obviously struggling at the initial low vision assessment.
The primary outcome will be the change in depressive symptoms, from randomisation to 6
months, as measured by the Becks Depression Inventory (BDI-II). The BDI-II is a reliable, valid and widely used outcome measure for depression [16
Secondary outcome measures in this exploratory trial will include change (from randomisation to 3 and 6
months) in visual disability as measured using the 7 item NEI VFQ and the near vision subscale of the VFQ-48 and, change in ‘generic health related quality of life’ as measured using the EQ-5D [18
]. Additional secondary outcomes measured at 6
months will include the proportion of people still screening positive for depression as measured with the GDS-15, participant’s use of services as determined using a Client Services Receipt Inventory (CSRI) and an economic measure of ‘capability and wellbeing’ the ICECAP-O [20
]. At the end of this exploratory trial participants will be asked about any treatments for depression they have received during the trial period (specifically, whether they received regular review and support by their GP, referral to a community mental health team, supportive psychotherapy, counselling, problem-solving therapy, cognitive-behavioural therapy or antidepressant therapy). They will also be asked about the acceptability of any interventions received. Participants in the waiting list control group who still screen positive for depression (GDS-15, 6+) will be referred to their GP. And, 3
months later asked about any treatment for depression received as a result of the referral.
All trial outcomes will be assessed by telephone interview by the Study Coordinator who will be masked to the treatment intervention. In order to minimise ‘loss to follow up’, interviews will be based on a standard script and will remind participants of their valuable contribution and the need to complete the study.
During the course of the study 1000 consecutive attendees will be screened. At the outset we anticipate that about 1/3 will screen positive for depression and be eligible, and that of those, about 100 will consent to take part. This target of 100 patients recruited into the trial is realistic and achievable. Resource limitations require us to cap the number of participants to 150.
Since this study is exploratory we will predominantly use descriptive methods to report results. That is, this study is not powered to look at differences in outcomes (there is no data upon which to determine the sample size). However, to give the reader some indication about the power of this exploratory trial, if we assume that those entering the trial are moderately depressed (average BDI-II score of 30, SD 10.5) and, if the trial finishes with only 20 participants per treatment group, we would have 83% power to detect a difference in post treatment means of 10 (a change of 10 units corresponding to a moderately important clinical difference) with a 5% two-sided significance level.
Participants will be identified and recruited from 7 community based low vision rehabilitation centres in South Wales and from one large secondary care based low vision rehabilitation centre at Guys and St Thomas’ NHS Trust, London [23
]. Based on known activity in these recruitment centres we anticipate that the 7 sites in Wales will screen approximately 500 people in total and that a similar number will be screened in London.
Adults (18+ years of age) with a score of at least 6 on the GDS-15 (administered as part of the routine pre assessment questionnaire) will be the inclusion criterion for this study.
To ensure our results are applicable to the population we propose to keep the exclusion criteria to a minimum, specifically: 1) those who have already had a low vision assessment within the previous 12
months, 2) those who are referred to the clinic in error (i.e. so visually impaired that low vision care and follow up are inappropriate or, so visually able (e.g. following refraction) that there are no low vision needs), 3) those already being treated for depression including psychotherapeutic and psychopharmacological treatments, 4) inability to understand English, 5) Inability to use the telephone e.g. caused by very poor hearing, 6) severe medical illness that would preclude participation in a 6
month study, 7) a score of 2 or 3 on the BDI-II question about suicidal ideation, all people in this group will be urgently referred to their GP, 8) those screening positive for significant cognitive / memory problems as determined with a modified version of the MMSE will be excluded.
Trial procedures: consent, baseline assessments and randomisation
The low vision rehabilitation services in Wales and London now include a routine self-complete pre assessment questionnaire (made up of the 7 item NEI VFQ and GDS-15) which is given to patients one week before their consultation. During the trial this will be accompanied by a Patient Information Sheet letting people know that the low vision service they are about to attend is taking part in a study. The sheet details the aims, purpose and what is involved in participation in the study. At the rehabilitation consultation, those satisfying the inclusion criteria will be invited to take part by the optometrist who provides the low vision rehabilitation intervention.
The optometrists (who are trained to take informed consent) will document the consultation using standard clinical records and, for the purposes of this study, a case report form. The GDS-15 will be reviewed at the start of the hour long low vision assessment and all positive screening patients will be informed that they may be eligible to take part in the study. Toward the end of the low vision assessment a full verbal explanation of the study will be provided to positive screening patients and they will be asked if they would like to take part. Those expressing an interest in taking part will then be screened by the optometrist for exclusion criteria 1, 2, 3, 4, 5 & 6. Those still eligible will be given written copies (large print) of the questionnaires that will be used in the telephone interview to take home and will be asked to provide their written informed consent.
Consenting patients will be contacted by the Study Coordinator within 1
week to, make an appointment to administer the cognitive screen and BDI-II to check for the exclusion criteria 7 and 8. Those meeting the additional eligibility criteria will be enrolled in the study and the other baseline outcome measures will be administered. Those scoring 29 or more on the BDI-II at the baseline interview have severe depressive symptoms and in all cases a letter will be written to the person’s GP requesting a ‘medication evaluation’.
Potential recruits who would like additional time to consider enrolment, after the initial low vision assessment will be contacted by phone by their low vision practitioner one week later to find out “how they are getting on”. If they then want to take part in the study they will be booked back in to see the practitioner to complete the study consent form. Participants who screen positive for depression but who decline to take part in the trial will be asked if they would like to be referred to their GP anyway. The person’s decision will be noted in the records e.g. if they decline referral. These people will be asked if they are willing to provide their consent for the use of anonymous baseline data.
Within one week of the baseline telephone interview, participants will be randomised using computer generated permuted blocks of varying sizes (unknown to investigators) stratified for study centre (Wales and London) and for ‘medication evaluation’ only. To ensure that the Study Coordinator is masked to treatment allocation, the Chief Investigator will allocate participants to the different arms of the trial.
For those assigned to the ‘referral to GP’ arm of the trial, the letter to the GP will be posted on the day the person was randomised and the participant will be asked to make an appointment with their GP within 2
weeks. For those assigned to PST, treatment will commence within 2
weeks of randomisation. All participants will be given a follow up vision rehabilitation appointment 6
weeks after the initial rehabilitation appointment.
The Study Coordinator will contact participants 3 and 6
months after randomisation to obtain outcome measure data.
A process evaluation will be conducted to evaluate whether the intervention was carried out in accordance with the trial protocol. Specifically, all PST consultations will be taped and 1/3 of them will be reviewed by the research team to ensure the fidelity of this intervention. The fidelity of the ‘referral to GP’ arm will be determined by noting the date the referral letter was sent to the GP. After the 6
month outcome measure a letter will be sent to each participant’s GP asking what treatments / action have taken place during the trial period. At the end of the trial all participants will take part in an interview to determine their level of satisfaction with the service received and any action taken by the GP. This final interview will include qualitative questions about what worked well and not so well.
Assessment of costs
We will also conduct an assessment of the direct costs associated with the delivery of the PST intervention and identify its key cost consequences e.g. reductions in demands on other service. The cost of the intervention will be assessed by monitoring the activity of the Psychological Therapist and the costs associated with travel to and from participants’ homes.
The costs associated with the use of additional services will be established using a telephone based Client Service Receipt Inventory (CSRI) [21
We will tabulate the number of patients by GDS-15 Score. We will estimate the prevalence of depressive symptoms (defined as a GDS-15 score of 6 or more) with 95% confidence intervals computed by the exact binomial method. We will use logistic regression to assess whether factors such as age, sex, pathology and site are associated with depressive symptoms. With a sample size of 1000, a two sided 95% confidence interval will extend 0.028 from the observed proportion for an expected proportion of 0.3.
This trial is the first of its kind and hence ‘exploratory’. For the trial, we will compare baseline characteristics of the allocated groups to ensure the adequacy of randomisation (including duration of visual loss). If outcome data are not available for all randomised patients, we will compare characteristics of those for whom it is available with those for whom it is not to assess whether the groups differ systematically. To retain the validity of the randomisation process the analyses will be undertaken on an intention-to-treat basis.
We will report average change in BDI scores in treatment groups using the mean or median as appropriate together with some indication of variability such as the standard deviation or interquartile range. These results will then be used to estimate the numbers of patients required for a definitive study.
Missing data has the potential to bias our findings. Hence we will attempt to minimise ‘loss to follow up’ by making it clear to potential recruits what is required at the outset, letting them know that completion of the study is important and reminding them of the value of their contribution when outcomes are assessed during the scripted telephone interviews. We will document the reasons for missing data and establish both the extent and patterns of missing data in each arm of the trial. We will also acknowledge the potential impact of missing data in our conclusions.
We will also compare the characteristics (age, sex, level of depression, level of vision impairment, ethnicity, functional ability and time since vision loss first identified) of those who did and did not consent to take part in the trial to determine if we can predict who is likely to benefit from depression screening.
We will also look for PST trainer effects (i.e. compare outcomes of PST delivery in London and Wales). The study design and scripted outcome measurements are designed to minimise the possibility of masking violation in this exploratory trial. We will however report any masking violations descriptively.
Analyses of costs
We will determine the direct costs of the psychological problem solving based intervention for people experiencing sight loss. We will also determine preliminary data on the incremental cost effectiveness of a home-based problem solving psychological intervention, aimed to support people experiencing sight loss, compared to referral to primary care and a waiting list control.
From a public sector, multi-agency perspective we will: fully cost the development, staff training and roll-out of this problem-solving psychological intervention [24
]. We will record study participant primary and secondary care health service use, social care use and contacts with voluntary sector agencies supporting blind and partially sighted people (using an interviewer administered, telephone-based Client Service Receipt Inventory (CSRI), as one of the full battery of study measures, costed using National unit costs and where necessary, local costs). We will also conduct a primary cost effectiveness analysis (using the Geriatric Depression Scale (GDS-15) as our measure of effectiveness); conduct a secondary cost-utility analysis using EQ-5D (interviewer administered by telephone) as our measure of utility to generate a cost per QALY (ICER) and Cost Effectiveness Acceptability Curve (CEAC) for comparison with the NICE ceiling of £30,000 [25
As this is a feasibility study we will pay particular attention to the validity of EQ-5D in this group of patients, comparing the 5 domains of EQ-5D (including anxiety and depression) with, in particular, the depression measures used in this study [28