ORIGINAL ARTICLE

Contribution of research networks to a clinical trial ofantidepressants in people with dementia

NIALL MCCRAE1, LISA DOUGLAS2 & SUBE BANERJEE3

1Florence Nightingale School of Nursing & Midwifery, King’s College London, Room 1.17, JamesClerk Maxwell Building, London, UK, 2MHRN North West Hub, St Catherine’s Hospital,Birkenhead, Wirral, UK, and 3Health Services & Population Research Department, Institute ofPsychiatry, King’s College London, London, UK

AbstractBackground.: The Mental Health Research Network and Dementia and Neurodegenerative DiseasesResearch Network were established in the UK to increase research capacity and activity; the formerin mental health generally, and the latter specifically in neurodegenerative disorders including demen-tia. Little evidence exists on the impact of these networks on research in mental health of older people.Aims: To examine research network support to a multi-centre randomised controlled trial of anti-depressants in people with depression superimposed on dementia.Method: Qualitative questionnaires were completed by principal investigators, trial-funded researchworkers and clinical study officers (CSOs) of the research networks.Results: Research network contribution was much valued by principal investigators and the nineresearch teams. CSOs boosted the recruitment campaign in a challenging environment and enhancedassessment processes. Some problems with consistency and staff turnover were raised.Conclusion: Research network input can make an appreciable difference to the process and outcome of amulti-centre clinical trial.

Keywords: research networks, trial management, mental health, recruitment, depression, dementia

Background

With an ageing population and consequent increase in the prevalence of dementia and otherpsychological and neurological disorders associated with old age, the mental health of olderpeople is steadily progressing from an erstwhile “Cinderella” status to a priority for researchand development. However, due to an inadequate research infrastructure in the UK, theunique advantages of the National Health Service (NHS) as a setting for treatment evaluationhave not been fully exploited. It is widely believed that the most definitive evidence is producedby experimental investigation, in the form of the randomised controlled trial (RCT).Until recently, mental health trials in the NHS have tended to be small and local, thus

Correspondence: Niall McCrae, Florence Nightingale School of Nursing & Midwifery, King’s College London, Room 1.17, JamesClerk Maxwell Building, 57 Waterloo Road, London SE1 8WA, UK. Tel: +44 20 7848 3619. E-mail: n.mccrae@kcl.ac.uk

Journal of Mental Health, 2012; 21(5): 439–447© 2012 Informa UK, Ltd.ISSN: 0963-8237 print / ISSN 1360-0567 onlineDOI: 10.3109/09638237.2012.664298

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

underpowered and of limited generalisability (Medical Research Council, 2010). The reportStrengthening Clinical Research (Academy of Medical Sciences, 2003) described a series ofshortcomings in research activity and capacity across the spectrum of healthcare in the UK,but highlighted mental health as one of the poorest areas of development. The UK ClinicalResearch Collaboration (2006), founded in 2004 by the Department of Health to boost theclinical research capacity, reported that mental health received only 6.5% of fundingprovided by the major research bodies, compared to 25% allocated to cancer studies.

The success of concerted action to develop cancer research led to the formation of nationalresearch networks for other healthcare priorities, funded by the Department of Health underthe umbrella organisation of the National Institute for Health Research. These networksprovide focused investment in research within the NHS, engaging academic experts,clinicians, patients and their carers, and biotechnology and pharmaceutical industries in acollaborative enterprise. In 2003, the Mental Health Research Network (MHRN) wasestablished to provide support for the large, multi-centre trials necessary to inform policyand practice. The Dementia and Neurodegenerative Diseases Research Network(DeNDRoN) was founded in 2005 to facilitate research in dementia and other organicbrain disease. Both networks are funded by the National Institute of Health Research.

The MHRN and DeNDRoN have regional structures comprising clinical and academicorganisations, bringing together a wide range of skills and expertise in research and practice.Once a study is formally adopted, clinical study officers (CSOs) are deployed on a sessionalbasis (there are workers with other job titles relating to qualifications, but for brevity the term“CSO” is used generically in this paper). The role of the CSO is primarily concerned withachieving recruitment targets, but other tasks include applying for local research approvals,liaison with clinical teams, obtaining consent and conducting assessments. A recent MedicalResearch Council (2010) review regarded these networks as a substantial boost to mentalhealth research in the UK. However, while the experience and impact of research networkinvolvement have been studied in other fields such as cancer (Cox et al., 2005) and diabetesmellitus (Kennedy et al., 2010), there is currently little evidence in mental health. Despitethe increasing contribution of research networks in mental health studies, little evidencehas been published on their input beyond acknowledgements of support (Ersche et al.,2010; Rathod, 2010; Slade et al., 2010).

The trial

HTA-SADD (Health Technology Assessment: Study of Antidepressants for Depression inDementia) is a large multi-centre RCT of the clinical and cost effectiveness of antidepressantmedication for depression in people with Alzheimer’s disease, funded by the Health Techno-logyAssessment programme.Depression is a common condition in later life and is particularlycommon in dementiawith a prevalence ofmajor depressive disorder of up to 22% and 27% forminor depression (Lee & Lyketsos, 2003). As coexisting depression exacerbates behaviouraldisturbance in Alzheimer’s disease and carer stress, treatment is a clinical priority.

Antidepressant drugs are widely used to treat depression in people with dementia, despitea Cochrane review showing only weak evidence for such intervention (Bains et al., 2002).HTA-SADD was designed as a definitive RCT to determine short-term and long-term clini-cal effectiveness, benefit-to-harm ratio and cost-effectiveness of sertraline and mirtazapine,two commonly used antidepressants in the treatment of depression in dementia. Theprimary outcome measurement was the Cornell Scale for Depression in Dementia(CSDD; Alexopoulos et al., 1988). This is a validated tool for distinguishing symptoms ofdepression from co-existing dementia, administered to both the patient and primary carer,

440 N. McCrae et al.

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

followed by assessor’s rating. Multi-centre ethics approval was gained in June 2006, followedby site-specific approvals.

The trial ran at nine regional sites participated (Birmingham, Cambridge, Leicester,Liverpool, Manchester, Newcastle, North London, Southampton and South London).Recruitment began in January 2007, with all sites actively recruiting by May 2007 (a staggeredstart was due to delays in research governance approvals and in appointment of grant-fundedresearch workers). Old age psychiatrists were invited to refer eligible patients to a full-time localresearch worker, who screened patients with the CSDD (minimum score 8). Supervised by aprincipal investigator at each site, research workers conducted baseline and outcome assess-ments, and monitored compliance and safety. Baseline, week 13 and week 39 assessmentsincluded the CSDD, Client Service Receipt Inventory, mini-mental state examination, andscales measuring behavioural and psychological symptoms, quality of life and carer burden.In an amendment to the initial design, the recruitment period was extended from 12 to 30months, with the target of 507 participants reduced to 338. Ultimately, 326 patients wereenrolled. The trial results have been published elsewhere (Banerjee et al., 2011).

The MHRN and DeNDRoN made formal arrangements to support HTA-SADD. At theoutset, MHRN hubs sought the necessary approval from local research ethics committeesand research governance departments of mental health trusts. When sites extended intoadditional mental health trusts, it was normally the MHRN that managed the submissions(an integrated ethics and research governance application was introduced during the trial,making this process easier). As trial sites expanded, CSO deployment increased. Therewas no uniformity of research network input; instead, this varied between sites throughoutthe trial, depending on organisational factors within the networks, staff availability andturnover, and local policy on the CSO role. Some sites benefited from substantial supportfrom both MHRN and DeNDRoN, while others had periods without any CSO.

Overall, 54 CSOs were allocated to the trial, with highly variable levels of input (some pro-vided little or no time to HTA-SADD due to redeployment to other studies, or leaving theirposts). As an approximate guide, sites were allocated at least one CSO in the host mentalhealth trust (contributing to screening, and in many cases to baseline and prospective assess-ments), with two CSOs in other trusts (mainly for screening). We estimate the overall inputCSOs as equivalent in time to that of the grant-funded research worker. The experience andimpact of this contribution were examined by this study, to inform future collaboration withresearch networks in clinical trials.

Method

Investigators and research staff were approached for a survey enquiring into their experiences ofresearch network involvement in HTA-SADD. A semi-structured qualitative questionnaire wasdesigned for each of three groups: principal investigators, grant-funded research workers andresearch network CSOs (see appendices). Approval to approach CSOs was obtained fromthe MHRN and DeNDRoN. Many of the CSOs had left their research network posts, butthose remaining in contact with the research team were pursued. The survey was conductedon the website www.surveymonkey.com, and only N.M. and L.D. had access to responses.

Results

Participants

Table I summarises recruitment to the survey.

Contribution of research networks to RCT 441

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

Five of the eight trial-funded research worker respondents were in post at site initiationand seven were in post at time of the survey; average duration in the trial was over 3 years.Research network respondents had spent a mean of 16 months working on the trial, witha range from 6 to 30 months. The background of trial-funded research workers and researchnetwork CSOs is presented in Table 2.

Initial expectations

For most of the principal investigators, this was their first experience of working with the re-search networks. There had been initial uncertainty about the collaboration, but the mostcommonly anticipated input was with recruitment. Only two principal investigators had pre-dicted an ongoing assessment role for CSOs. Although the research ethics and governanceapplications at each site were prepared by the MHRN, this was not mentioned by the prin-cipal investigators. Most of the trial-funded research workers had little awareness of the re-search networks at appointment and had been unsure about whether research networksupport would be limited to screening or include drug delivery and outcome assessments.

Research network staff: general role

Role expectations of research network CSOs varied by background. Psychology graduatestended to see the job as a stepping stone to a clinical psychology course, or to PhD enrol-ment. This was apparent in their interest in gaining experience in multiple studies, whilenurses were more inclined towards intensive contact with participants. The research net-works offered nurses an opportunity to engage in research, an activity to which they havelimited access in clinical practice. They were enthused by contributing to the evidencebase, and appreciated the stimulating environment of the research network:

It was incredibly refreshing to move from a clinical setting where you were controlled andregulated to an environment where you are encouraged to think!

Table I. Participants.

Group Approached Responded (%)

Research network CSOs 31 20 (65%)Trial-funded research workers 11∗ 8 (73%)Principal investigators 13† 8 (62%)Total 55 36 (65%)

∗One site had three successive trial-funded research workers.†Four consultant psychiatrists with substantial input to trial conductand management were also included.

Table II. Background of research workers.

Background Research network staff ∗ Trial-funded research workers

Nursing 7 4Psychology graduate 12 3Other graduate 1 1

∗1 missing datum.

442 N. McCrae et al.

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

Dissatisfying aspects of the CSO role included a perceived piecemeal involvement in studiesand consequent lack of ownership. Being judged on recruitment performance was a problem,as there was often little reward from attending meetings or screening medical records, and itwas felt that positive results were credited to the main research worker. Lack of clear roledemarcation was observed by some research network respondents:

We did not have a clear remit as to what we would be expected to do on a study, and attimes this was open to interpretation.

Research network staff: experience with HTA-SADD

Two-thirds of the research network respondents were allocated up to 1 day per week on thetrial, with a third contributing 2–3 days per week. These were average amounts of input; gen-erally, HTA-SADD time reduced as new studies were added to the local network portfolio.

I was allocated one full day to HTA-SADD initially. After the recruitment improved andmy links with services were well established, I reduced my time to half a day. The remain-der of my time was split between many studies – at one point I was working on nine studies.Thankfully not all of them required me to have such a “hands-on” role, which is why I wasable to balance and organise my time.

Although the main thrust of the CSOs was in boosting recruitment, level of input in HTA-SADD ranged from identifying potential referrals to the full set of tasks performed by thetrial-funded research worker; this varied not only between sites, but also between CSOs.Most research network respondents had experienced HTA-SADD as one of their moreinteresting studies. They enjoyed developing a rapport with clinicians, and working in thespecialised field of dementia. For CSOs restricted to screening activity, the HTA-SADDexperience was sometimes frustrating due to the difficulty in persuading clinicians to referpatients. One CSO expressed regret at not being involved beyond identifying potential cases:

I never did get to carry out many of the patient assessments, which was disappointing as Iwould have liked to do more of those – it would have been a nice reward as focusing onrecruitment alone can be draining.

While most respondents liked working with the HTA-SADD study teams, a minority did notfeel valued, or perceived that their contribution was constrained by the trial-funded researchworker, sometimes causing tension. However, positive comments were made about theorganisation of the trial, which included regular meetings for research staff in London, towhich all CSOs were invited.

The bi-yearly dinner and meeting was a really great way of fostering enthusiasm and drive –none of the other studies I worked in had this.

Site staff experience of research network input

As one principal investigator stated, this was “a new experience for all of us”. Principal inves-tigators were generally pleased with research network support, which increased the capacityfor recruitment over multiple trusts. In the host sites, the presence of research network staffwas evident in their role in screening and in assessment; at additional sites, recruitment oftendepended entirely on research network staff. Positive observations included the CSOs being

Contribution of research networks to RCT 443

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

based in the local NHS trust, good communication, and interchangeability with the mainresearch worker.

Excellent commitment and support from DeNDRoN, not just with recruitment but allaspects of study including providing cover for absences.

Although views varied with the level of input, most of the trial-funded research workersvalued the support from the research networks. Particularly appreciated was the opportunityat some sites for routine assessment in pairs, which allowed the carer to speak to the researchworker in private, while a CSO conducted the other part of the assessment with the patient(or vice versa). On solo visits, it was not always possible to conduct the assessments separatelydue to the behavioural symptoms of dementia. Research network input thus enhanced thequality of assessment. It also provided emotional support to research workers who otherwiseworked alone. Help with recruitment, data entry and drug delivery was also acknowledged.

I liked the fact that the research networks were flexible and willing to adapt to the needs ofour site, and were able to expand the remit of their research workers to meet our require-ments. I also found the majority of their research workers were enthusiastic, knowledgeableand hard-working.

However, there was also criticism by the main research workers of increasing restrictions onthe CSO role by network management. A perpetual problem was staff turnover in theresearch networks, with consequent lack of continuity and dedication. This caused gaps insupport, and a recurring need to build new working relationships. Three principal investi-gators criticised a lack of consistency in CSO allocation, with a perception that someworkers lacked adequate knowledge of the trial.

Suggestions for improvement

Research network respondents made suggestions to enhance their contribution to clinicaltrials such as HTA-SADD. These included stronger connections with the principal investi-gator and clinicians, better awareness of the scope of research network activity and clearerrole demarcation with trial-funded research workers. Trial-funded research workerssuggested that a more fulfilling involvement in trials would improve job satisfaction forCSOs, thereby perhaps reducing turnover. Principal investigators urged clarity and consist-ency in the support provided for trials, some arguing that research network staff should bedecentralised and integrated in study teams.

They should be directly attached to the study and take part in the psychometric testing toembed them in the trial and give them ownership.

Indeed, two principal investigators proposed a contractual commitment. By contrast, otherprincipal investigators were aware that by having CSOs based within mental health trusts, theresearch networks enabled sites to expand over awider geographical area than initially planned.

Discussion

This study indicates that the research networks can make a valuable contribution to theimplementation of a large, multi-centre clinical trial. In HTA-SADD, the MHRN provided

444 N. McCrae et al.

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

an important service in gaining local ethics committee and NHS trust approvals, which canbe a time-consuming process. This was apparently not fully acknowledged by the site teams,for whom a more visible demonstration of research network support was in the conduct of thetrial, while the activity around gaining approvals was perhaps taken for granted. As well asboosting a recruitment campaign in a challenging environment, many CSOs contributedto the monitoring and assessment of participants. Particularly useful in the context of atrial in people with dementia was the facility to conduct assessments in pairs. Trials relyon accurate data, and enabling people with dementia and their carers to speak privatelyfrom each other, and thus more candidly, almost certainly enhanced the quality of data col-lection. These two outcomes would justify research network input to clinical trials, but therewere interesting differences between sites and between workers.

As well as the tangible outcomes of network collaboration, the survey focused on job sat-isfaction among CSOs. This is an important matter not only for the networks but also forthe conduct of a large clinical trial, particularly in a study dealing with people with demen-tia and their carers. Morale and turnover of workers are likely to have an impact on therelationship between the trial team and participants. A sensitive rapport built on trust en-hances the quality of assessment, and the disruptive effect of having to divulge privatedetails to different workers could contribute to attrition rates in a trial such as HTA-SADD. This study revealed the extent to which CSOs felt engaged and rewarded in con-tributing to the trial.

Most sites benefited from support from both networks, and while recruitment and assess-ment activity were coordinated by the trial-funded research worker, there were potential pro-blems of duplication or unclear roles and responsibilities. Such difficulties may have beenexpected with an evolving organisational innovation, but were mostly resolved within thesites. There was also a degree of unrealistic expectation from principal investigators, someof whom wanted more structured commitment from the networks. Similar issues wereobserved with CSOs in the Cancer Research Network (Cox et al., 2005). New rolesrequire effective management to ensure that prescribed roles are preserved, without stiflingcreativity. During the recruitment period of HTA-SADD, the research networks were stea-dily developing their working systems, while CSOs were working on an increasing number ofstudies. Inevitably administrative procedures were introduced and limits were placed on CSOinput. Ng and Weindling (2009) observed that while the networks have undoubtedly sup-ported research in a complex regulatory environment, they have also added to the bureau-cratic burden. In HTA-SADD, the networks were generally found to be accessible andenthusiastic partners, and it must be hoped that organisational structures and processeswill facilitate rather than impede such collaborative ventures.

Variation between the input of CSOs was not only due to differences in prescribed role butalso their individual skills and inclinations. For CSOs aiming at a career in clinical psychol-ogy, the research networks provided a good opportunity to enhance their curriculum vitae,and HTA-SADD was a major trial in a priority area. In completing HTA-SADD assess-ments, CSOs gained experience with people with dementia and in a wide range of psycho-metric instruments. This was undoubtedly beneficial to the several CSOs who gained aplace in clinical psychology training during the trial. Alongside such highly motivatedyoung graduates, the research networks also employ nurses seeking a new challenge. Theirclinical experience is an asset to the conduct of a trial, particularly where participants arementally or physically impaired, or lacking capacity. A balance of psychology graduatesand nurses appears to be a strength of the networks. This survey shows that for someCSOs, particularly nurses, a more intensive role with plenty of participant contact is morefulfilling than the relatively unrewarding task of database screening.

Contribution of research networks to RCT 445

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

The limitations of this study should be considered. The response rate of two-thirds is notuntypical of this type of survey, but respondent bias is possible. Participation may have cor-related with level of involvement in the trial, reducing the representativeness of findings. Theroles of the authors in the clinical trial may have influenced responses. The product of CSOinput was not quantified, but it should be acknowledged that a figure such as number of re-cruits is not a straightforward measurement. Only a small proportion of referrals from psy-chiatrists resulted directly from case-finding by research workers, whose presence at clinicsand meetings may have stimulated referrals, but in an unquantifiable way. Trial recordsshowed which psychiatrist referred a patient rather than who screened the notes. We rec-ommend further research to pursue meaningful activity and outcome data, making judicioususe of information collected routinely by network administration. Such empirical data wouldadd to the experiential findings reported here. Qualitative interviews would provide richer ac-counts than permitted by an online questionnaire. This article describes only one study: asexperiences are likely to vary, findings from other mental health trials would contribute tomore definitive evidence on research network support.

Conclusions

With challenging recruitment targets and large geographical areas to cover, research networksupport can make the difference between success and failure in clinical trials. A continualprocess of evaluating research network support will contribute to optimising the value ofthis resource.

Acknowledgements

The authors would like to thank all respondents for their contribution to the study. JoanneAshcroft (deputy assistant director, MHRN) gave valued advice on the questionnairedesign and draft paper. Views expressed are the authors’ own rather than those of the researchnetworks.

Declaration of Interest: HTA-SADD was funded by the UK Department of Health, butthe study reported here received no additional funding.

Conflict of Interest Statement: N.M. was the trial manager of HTA-SADD. He led thedesign of the study reported here, performed data collection and analysis, and wrote thepaper. L.D. was the grant-funded research worker in Liverpool and latterly held a managerialpost in MHRN North West hub. She helped to design the questionnaires and analyse thedata from CSO and PIs. She was also invited to complete the grant-funded researchworker questionnaire, but did not have access to the data from others in this role. S.B. wasthe chief investigator of HTA-SADD. He supervised the design and conduct of the studyand commented on the draft paper. He was invited to complete the principal investigatorquestionnaire as local investigator of the South London site, but did not have access to thedata from any other participants.

References

Academy of Medical Sciences. (2003). Strengthening clinical research. London: Academy of Medical Sciences.Alexopoulos, G.S., Abrams, R.C., Young, R.C., & Shamoian, C.A. (1988). Cornell Scale for Depression in

Dementia. Biological Psychiatry, 23, 271–284.

446 N. McCrae et al.

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.

Bains, J., Birks, J.S., & Dening, T.R. (2002). Antidepressants for treating depression in dementia. Cochrane DatabaseSystematic Reviews, 4, CD003944.

Banerjee, S., Hellier, J., Dewey, M., Romeo, R., Ballard, C., Baldwin, R., et al. (2011). Sertraline or mirtazapine fordepression in dementia (HTA-SADD): A randomised, multicentre, double-blind, placebo-controlled trial.Lancet, 378, 403–411.

Cox, K., Avis, M., Wilson, E., & Elkan, R. (2005). An evaluation of the introduction of clinical trial officer roles intothe cancer clinical trial setting in the UK. European Journal of Cancer Care, 14, 448–456.

Ersche, K.D., Turton, A.J., Pradhan, S., Bullmore, E.T., & Robbins, T.W. (2010). Drug addiction endopheno-types: Impulsive versus sensation-seeking personality traits. Psychiatry, 68, 770–773.

Kennedy, E.D., Chester, P., Majeed, A., Wallace, P., & Matthews, D.R. (2010). Networks offer new opportunitiesfor diabetes research. Primary Care Diabetes, 4, 3–8.

Lee, H.B., & Lyketsos, C.G. (2003). Depression in Alzheimer’s disease: Heterogeneity and related issues. BiologicalPsychiatry, 53, 354–362.

Medical Research Council. (2010). Review of mental health research: Report of the strategic review group. London:Author.

Ng, S.M., & Weindling, A.M. (2009). The impact of networks on clinical trials in the United Kingdom. Journal ofResearch in Nursing, 15, 203–211.

Rathod, S. (2010). Developing culturally sensitive cognitive behaviour therapy for psychosis for ethnic minoritypatients by exploration and incorporation of service users’ and health professionals’ views and opinions.Behavioural & Cognitive Psychotherapy, 38, 511–533.

Slade, M., Byford, S., & Barrett, B. (2010). Alternatives to standard acute in-patient care in England: Short-termclinical outcomes and cost-effectiveness. British Journal of Psychiatry, 197, S14–S19.

UK Clinical Research Collaboration. (2006). UKHealth Research Analysis. Retrieved February 22, 2012, from www.ukcrc.org/researchcoordination/healthresearchanalysis/ukanalysis/

Appendix I: Questionnaire for CSOs

(1) What attracted you to work for the research network?(2) State your background: nursing/psychology/other (please specify).(3) What did you like most about the CSO role, generally?(4) What did you like least about the CSO role, generally?(5) Please state time period of your contribution to the trial.(6) Initially, what was your planned role with HTA-SADD?(7) During the HTA-SADD recruitment phase, how was your time divided between studies? (in a typical week)(8) What have you found satisfying with HTA-SADD?(9) What have you found dissatisfying with HTA-SADD?(10) How would you change or enhance the CSO role in studies such as HTA-SADD?

Appendix II: Questionnaire for trial-funded research workers

(1) Please state the period of your employment on HTA-SADD.(2) State your background: nursing/psychology/other (please specify).(3) Briefly describe the involvement of research network(s) in HTA-SADD at your site.(4) What were your initial expectations about CSO support?(5) What did you like most about research network/CSO input?(6) What did you like least about research network/CSO input?(7) How would you change or enhance the CSO role in studies such as HTA-SADD?

Appendix III: Questionnaire for principal investigators

(1) Briefly describe the involvement of research network(s) in your area, during the HTA-SADD recruitmentphase.

(2) What was your initial expectation of research network input with HTA-SADD?(3) What have you found satisfying with research network input?(4) What have you found dissatisfying with research network input?(5) How would you change or enhance research network input in studies such as HTA-SADD?

Contribution of research networks to RCT 447

J M

ent H

ealth

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f U

lste

r at

Jor

dans

tow

n on

11/

28/1

4Fo

r pe

rson

al u

se o

nly.