[Show abstract][Hide abstract] ABSTRACT: Researchers have seldom included bereaved parents in studies of participants' views of randomised controlled trials (RCTs); hence our understanding of the impact of trials is based on skewed and incomplete samples. Little is known about parental experiences of the death of a child subsequent to their enrolment in a trial or of provision made for this experience by clinicians and trial teams. The Bereavement and RAndomised ControlLEd Trials (BRACELET) study was funded to consider bereavement in the context of paediatric intensive care (PIC) and neonatal intensive care (NIC) trials.
DESIGN AND METHODS The study comprised three interlinked components: a quantitative survey of RCT activity in UK paediatric intensive care units (PICUs) and neonatal intensive care units (NICUs), UK RCT recruitment and mortality rates, and provision for bereavement during 2002-6; a qualitative interview study involving 51 bereaved parents and 59 clinicians and trial team members associated with five neonatal trials; and a methodological study to inform future research.
RESULTS Fifty RCTs were identified as having enrolled babies or children from 2002 to 2006. Approximately 50% of UK NICUs and PICUs (54 NICUs, six PICUs) participated in at least one of these trials. Collectively they enrolled over 3000 children. Most enrolled small numbers, the majority of participants being enrolled by a small group of academic medical units. The proportion of deaths following trial enrolment was 17% in NIC trials and 6% in PIC trials. The qualitative study showed that trial-related decisions were made in a range of circumstances, some after extremely preterm births, others after complicated term deliveries, often under time pressures and in escalating crises. Parents' interest in trials appeared to recede initially but could re-emerge over time. They often valued opportunities to engage with a trial and were interested in more contact and information than they actually received. Clinicians often saw NICU bereavement policies as meeting parental needs, and trial participation as being of relatively minor significance in bereavement. This view may result from the positioning of clinicians' encounters with parents only in the initial stages of grief when trials were not a priority. Trial teams used a range of bereavement strategies, from no further contact to a pioneering multipart follow-up package. Communication with bereaved parents was complicated by limited contact opportunities. Trial teams were obliged to work without knowing whether their communications were appreciated, were problematic, or even whether they were received by parents. The methodological component highlighted strategies for recruitment and data collection in this sensitive setting. Recruitment by unsupported postal contact generally failed and a more personal approach via clinicians was more effective, supplemented by publicity material distributed via trusted organisations.
CONCLUSIONS A co-ordinated response to bereavement is as much a part of the running of trials as recruitment, and needs to be considered at trial inception. BRACELET has demonstrated the value and feasibility of research with bereaved parents involved in NIC trials. In order to respond to bereavement in a fair and sensitive way, as well as to better inform the design of RCTs, it is crucial that we listen to bereaved parents and evaluate new methods for so doing. More research is therefore needed into the experiences of bereavement subsequent to trial enrolment, with study of bereavement strategies in NIC trials as they are introduced. In addition, future studies should determine whether parents and triallists in PIC trials (and trials in adults) face the same issues as in NIC trials. Careful studies are necessary to explore how feedback of trial results are received and understood by bereaved and non-bereaved parents, and how individual trial teams manage this situation. An additional research area for exploring experiences of parenting twins and higher-order births in trials arose from BRACELET. Developmental research should continue to explore means of involving a wider range of parents in future research, including via publicity and specialist websites. Finally, methodological research is needed to ensure that we have the tools to explore, with parents and other relatives, as partners in research, a range of trial-related topics, which might be challenging, as the information is complex or the focus is sensitive.
FUNDING Funding for this study was provided by the Health Technology Assessment programme of the National Institute for Health Research.
No preview · Article · Jul 2014 · Health technology assessment (Winchester, England)
[Show abstract][Hide abstract] ABSTRACT: Early research in adults admitted to intensive care suggested that tight control of blood glucose during acute illness can be associated with reductions in mortality, length of hospital stay and complications such as infection and renal failure. Prior to our study, it was unclear whether or not children could also benefit from tight control of blood glucose during critical illness.
This study aimed to determine if controlling blood glucose using insulin in paediatric intensive care units (PICUs) reduces mortality and morbidity and is cost-effective, whether or not admission follows cardiac surgery.
Randomised open two-arm parallel group superiority design with central randomisation with minimisation. Analysis was on an intention-to-treat basis. Following random allocation, care givers and outcome assessors were no longer blind to allocation.
The setting was 13 English PICUs.
Patients who met the following criteria were eligible for inclusion: ≥ 36 weeks corrected gestational age; ≤ 16 years; in the PICU following injury, following major surgery or with critical illness; anticipated treatment > 12 hours; arterial line; mechanical ventilation; and vasoactive drugs. Exclusion criteria were as follows: diabetes mellitus; inborn error of metabolism; treatment withdrawal considered; in the PICU > 5 consecutive days; and already in CHiP (Control of Hyperglycaemia in Paediatric intensive care).
The intervention was tight glycaemic control (TGC): insulin by intravenous infusion titrated to maintain blood glucose between 4.0 and 7.0 mmol/l.
This consisted of insulin by intravenous infusion only if blood glucose exceeded 12.0 mmol/l on two samples at least 30 minutes apart; insulin was stopped when blood glucose fell below 10.0 mmol/l.
The primary outcome was the number of days alive and free from mechanical ventilation within 30 days of trial entry (VFD-30). The secondary outcomes comprised clinical and economic outcomes at 30 days and 12 months and lifetime cost-effectiveness, which included costs per quality-adjusted life-year.
CHiP recruited from May 2008 to September 2011. In total, 19,924 children were screened and 1369 eligible patients were randomised (TGC, 694; CM, 675), 60% of whom were in the cardiac surgery stratum. The randomised groups were comparable at trial entry. More children in the TGC than in the CM arm received insulin (66% vs. 16%). The mean VFD-30 was 23 [mean difference 0.36; 95% confidence interval (CI) -0.42 to 1.14]. The effect did not differ among prespecified subgroups. Hypoglycaemia occurred significantly more often in the TGC than in the CM arm (moderate, 12.5% vs. 3.1%; severe, 7.3% vs. 1.5%). Mean 30-day costs were similar between arms, but mean 12-month costs were lower in the TGC than in CM arm (incremental costs -£3620, 95% CI -£7743 to £502). For the non-cardiac surgery stratum, mean costs were lower in the TGC than in the CM arm (incremental cost -£9865, 95% CI -£18,558 to -£1172), but, in the cardiac surgery stratum, the costs were similar between the arms (incremental cost £133, 95% CI -£3568 to £3833). Lifetime incremental net benefits were positive overall (£3346, 95% CI -£11,203 to £17,894), but close to zero for the cardiac surgery stratum (-£919, 95% CI -£16,661 to £14,823). For the non-cardiac surgery stratum, the incremental net benefits were high (£11,322, 95% CI -£15,791 to £38,615). The probability that TGC is cost-effective is relatively high for the non-cardiac surgery stratum, but, for the cardiac surgery subgroup, the probability that TGC is cost-effective is around 0.5. Sensitivity analyses showed that the results were robust to a range of alternative assumptions.
CHiP found no differences in the clinical or cost-effectiveness of TGC compared with CM overall, or for prespecified subgroups. A higher proportion of the TGC arm had hypoglycaemia. This study did not provide any evidence to suggest that PICUs should stop providing CM for children admitted to PICUs following cardiac surgery. For the subgroup not admitted for cardiac surgery, TGC reduced average costs at 12 months and is likely to be cost-effective. Further research is required to refine the TGC protocol to minimise the risk of hypoglycaemic episodes and assess the long-term health benefits of TGC.
Current Controlled Trials ISRCTN61735247.
This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 18, No. 26. See the NIHR Journals Library website for further project information.
[Show abstract][Hide abstract] ABSTRACT: Evidence from the USA suggests that the home-based Family Nurse Partnership program (FNP), extending from early pregnancy until infants are 24 months, can reduce the risk of child abuse and neglect throughout childhood. FNP is now widely available in the UK. A new variant, Group Family Nurse Partnership (gFNP) offers similar content but in a group context and for a shorter time, until infants are 12 months old. Each group comprises 8 to 12 women with similar expected delivery dates and their partners. Its implementation has been established but there is no evidence of its effectiveness.Methods/design: The study comprises a multi-site randomized controlled trial designed to identify the benefits of gFNP compared to standard care. Participants (not eligible for FNP) must be either aged < 20 years at their last menstrual period (LMP) with one or more previous live births, or aged 20 to 24 at LMP with low educational qualifications and no previous live births. 'Low educational qualifications' is defined as not having both Maths and English Language GCSE at grade C or higher or, if they have both, no more than four in total at grade C or higher. Exclusions are: under 20 years and previously received home-based FNP and, in either age group, severe psychotic mental illness or not able to communicate in English. Consenting women are randomly allocated (minimized by site and maternal age group) when between 10 and 16 weeks pregnant to either to the 44 session gFNP program or to standard care after the collection of baseline information. Researchers are blind to group assignment.The primary outcomes at 12 months are child abuse potential based on the revised Adult-Adolescent Parenting Inventory and parent/infant interaction coded using the CARE Index based on a video-taped interaction. Secondary outcomes are maternal depression, parenting stress, health related quality of life, social support, and use of services.
This is the first study of the effectiveness of gFNP in the UK. Results should inform decision-making about its delivery alongside universal services, potentially enabling a wider range of families to benefit from the FNP curriculum and approach to supporting parenting.Trial registration: ISRCTN78814904.
[Show abstract][Hide abstract] ABSTRACT: To investigate parents' views about deferred consent to inform management of trial disclosure after a child's death.
A postal questionnaire survey was sent to members of the Meningitis Research Foundation UK charity, whose child had suffered from bacterial meningitis or meningococcal septicaemia within the previous 5 years. Main outcome measures were acceptability of deferred consent; timing of requesting consent; and the management of disclosure of the trial after a child's death.
220 families were sent questionnaires of whom 63 (29%) were bereaved. 68 families responded (31%), of whom 19 (28%) were bereaved. The majority (67%) was willing for their child to be involved in the trial without the trial being explained to them beforehand; 70% wanted to be informed about the trial as soon as their child's condition had stabilised. In the event of a child's death before the trial could be discussed the majority of bereaved parents (66% 12/18) anticipated wanting to be told about the trial at some time. This compared with 37% (18/49) of non-bereaved families (p = 0.06). Parents' free text responses indicated that the word 'trial' held strongly negative connotations. A few parents regarded gaps in the evidence base about emergency treatments as indicating staff lacked expertise to care for a critically ill child. Bereaved parents' free text responses indicated the importance of individualised management of disclosure about a trial following a child's death.
Deferred consent is acceptable to the majority of respondents. Parents whose children had recovered differed in their views compared to bereaved parents. Most bereaved parents would want to be informed about the trial in the aftermath of a child's death, although a minority strongly opposed such disclosure. Distinction should be drawn between the views of bereaved and non-bereaved parents when considering the acceptability of different consent processes.
[Show abstract][Hide abstract] ABSTRACT: OBJECTIVE: to explore women's and their partners' views of recruitment to emergency trials in severe postpartum haemorrhage (PPH). DESIGN: interview-based qualitative study. In semi-structured in-depth interviews, five recruitment options for a PPH trial in an emergency context were considered. SETTING: interviews were carried out in participants' homes. PARTICIPANTS: nine women who had experienced a severe PPH and six partners. FINDINGS: interviewees rejected three options; decision-making by women prior to delivery, and by partners and legal representatives at the time of the emergency. Preferred options were women making antenatal decisions about trial entry themselves, followed by doctors making decisions at the time of the emergency. CONCLUSIONS AND IMPLICATIONS FOR PRACTICE: recruitment options involving women and their partners at the time of an emergency were rejected. Antenatal decision-making raises logistical and ethical considerations for emergency trial teams. Further research is needed to address the possibility of antenatal decisions for emergency trials and to develop and assess supportive post-enrolment recruitment and information strategies which take into account the stressful context of clinical emergencies such as PPH.
[Show abstract][Hide abstract] ABSTRACT: OBJECTIVE: to explore how severe postpartum haemorrhage (PPH) and its management is experienced by women and their partners, and how they later view events. DESIGN: interpretive phenomenological qualitative study with semi-structured interviews. SETTING: in-depth interviews were conducted in participants' homes and focused on experiences of PPH in hospital and post-discharge. PARTICIPANTS: nine women who had experienced severe PPH and six partners. FINDINGS: this study demonstrates the stressful and emotional nature of severe PPH and extends the literature by considering partners' perspectives. Women and men had different experiences and information needs, but interviewees often shared a common desire for help to understand past events. A dominant theme of communication difficulties, and two subthemes, disempowerment and information-deprivation arose strongly from the data. Communication difficulties were understandable during the emergency but were frustrating and upsetting in postpartum care and the longer term. CONCLUSIONS AND IMPLICATIONS FOR PRACTICE: research is needed to develop better communication and supportive strategies, which might avoid the sense of disempowerment in this potentially vulnerable population. Different information and support needs for women and men should be considered whatever policies are implemented.
[Show abstract][Hide abstract] ABSTRACT: Speaking to stressed parents in difficult circumstances about the care and prognosis for their very sick newborn baby is one of the challenges involved in providing neonatal intensive care. The regularity of the task can allow clinicians to develop their personal style and to refine their skills in communication, especially as many discussions have standard features such as explanations of a condition, an environment, a piece of equipment. Highly variable contextual and inter-personal elements can, however, add unpredictable dimensions to the discussions which can be demanding, professionally and personally. When a baby is a candidate for neural rescue, there are several additional issues which can increase these demands on clinicians.The purpose of this chapter is to help clinicians who are thinking through issues involved with communication in these difficult situations. We draw upon accounts from parents of critically ill babies of their experiences of discussing hypothermic neural rescue with clinicians. These data are taken from our own research; we have conducted two studies of the views and experiences of parents and clinicians involved in studies of hypothermia and neuroprotection. Allmark and Mason carried out a qualitative study which focused on the continuous consent processes used in the TOBY trial [1,2]; data from their TOBY-QUAL study are indicated by [TQ]. Snowdon and Elbourne carried out a qualitative study of parent and clinician-researcher views of a pre-trial safety study of hypothermia and ECMO ; data from their Views of Hypothermia and ECMO study are indicated by [VHE].
[Show abstract][Hide abstract] ABSTRACT: The objective of the study was to explore parental experiences of being offered participation in a previous neonatal research study involving venepuncture. The method employed was a questionnaire-based exploration of parents' attitudes in those approached to participate in a study of term and preterm immunization responses (Preterm Immunisation Study [PREMIS]). We explored experience of the initial approach, knowledge of study, venepuncture and views on research ‘in general’. In all, 59% of families responded. Highest response rates were for those participating in PREMIS (87% term/69% preterm) and lowest in decliners (34% and 35%). Responding parents participating in PREMIS were well informed, positive about research and did not find the venepuncture problematic. Sixty percent of responding parents who declined PREMIS attributed their declining to the need for venepuncture. In conclusion, parents participating or declining a neonatal study involving venepuncture are different, but participating parents were well informed and seemed able to judge that participation was right for them such that in consenting families venepuncture itself is not problematic.
[Show abstract][Hide abstract] ABSTRACT: The subject of death and bereavement in the context of randomised controlled trials in neonatal or paediatric intensive care is under-researched. The objectives of this phase of the Bereavement and RAndomised ControlLEd Trials (BRACELET) Study were to determine trial activity in UK neonatal and paediatric intensive care (2002-06); numbers of deaths before hospital discharge; and variation in mortality across intensive care units and trials and to determine whether bereavement support policies were available within trials. These are essential prerequisites to considering the implications of future policies and practice subsequent to bereavement following a child's enrollment in a trial.
The units survey involved neonatal units providing level 2 or 3 care, and paediatric units providing level II care or above; the trials survey involved trials where allocation was randomized and interventions were delivered to intensive care patients, or to parents but designed to affect patient outcomes.
Information was available from 191/220 (87%) neonatal units (149 level 2 or 3 care); and 28/32 (88%) paediatric units. 90/177 (51%) eligible responding units participated in one or more trial (76 neonatal, 14 paediatric) and 54 neonatal units and 6 paediatric units witnessed at least one death. 50 trials were identified (36 neonatal, 14 paediatric). 3,137 babies were enrolled in neonatal trials, 210 children in paediatric trials. Deaths ranged 0-278 (median [IQR interquartile range] 2 [1, 14.5]) per neonatal trial, 0-4 (median [IQR] 1 [0, 2.5]) per paediatric trial. 534 (16%) participants died post-enrollment: 522 (17%) in neonatal trials, 12 (6%) in paediatric trials. Trial participants ranged 1-236 (median [IQR] 21.5 [8, 39.8]) per neonatal unit, 1-53 (median [IQR] 11.5 [2.3, 33.8]) per paediatric unit. Deaths ranged 0-37 (median [IQR] 3.5 [0.3, 8.8]) per neonatal unit, 0-7 (median [IQR] 0.5 [0, 1.8]) per paediatric unit. Three trials had a formal policy for responding to bereavement.
A substantial number of deaths after trial enrollment were identified, distributed over many trials and units. Few trial teams had responses to bereavement in place. Those with the largest numbers of deaths might be best placed to collaborate in developing and assessing responses to bereavement.
[Show abstract][Hide abstract] ABSTRACT: There is increasing evidence that tight blood glucose (BG) control improves outcomes in critically ill adults. Children show similar hyperglycaemic responses to surgery or critical illness. However it is not known whether tight control will benefit children given maturational differences and different disease spectrum.
The study is an randomised open trial with two parallel groups to assess whether, for children undergoing intensive care in the UK aged <or= 16 years who are ventilated, have an arterial line in-situ and are receiving vasoactive support following injury, major surgery or in association with critical illness in whom it is anticipated such treatment will be required to continue for at least 12 hours, tight control will increase the numbers of days alive and free of mechanical ventilation at 30 days, and lead to improvement in a range of complications associated with intensive care treatment and be cost effective. Children in the tight control group will receive insulin by intravenous infusion titrated to maintain BG between 4 and 7.0 mmol/l. Children in the control group will be treated according to a standard current approach to BG management. Children will be followed up to determine vital status and healthcare resources usage between discharge and 12 months post-randomisation. Information regarding overall health status, global neurological outcome, attention and behavioural status will be sought from a subgroup with traumatic brain injury (TBI). A difference of 2 days in the number of ventilator-free days within the first 30 days post-randomisation is considered clinically important. Conservatively assuming a standard deviation of a week across both trial arms, a type I error of 1% (2-sided test), and allowing for non-compliance, a total sample size of 1000 patients would have 90% power to detect this difference. To detect effect differences between cardiac and non-cardiac patients, a target sample size of 1500 is required. An economic evaluation will assess whether the costs of achieving tight BG control are justified by subsequent reductions in hospitalisation costs.
The relevance of tight glycaemic control in this population needs to be assessed formally before being accepted into standard practice.
[Show abstract][Hide abstract] ABSTRACT: Qualitative interview study.
Fifty-nine patients with a family history of cancer who attend a regional cancer genetics clinic in the UK were interviewed about their current and previous research experiences. Findings: Interviewees gave a range of explanations for research participation. These were categorised as (a) social--research participation benefits the wider society by progressing science and improving treatment for everyone; (b) familial--research participation may improve healthcare and benefit current or future generations of the participant's family; and (c) personal--research participation provides therapeutic or non-therapeutic benefits for oneself.
We discuss the distinction drawn between motives for research participation focused upon self (personal) and others (familial/social), and observe that personal, social and familial motives can be seen as interdependent. For example, research participation that is undertaken to benefit others, particularly relatives, may also offer a number of personal benefits for self, such as enabling participants to feel that they have discharged their social or familial obligations. We argue for the need to move away from simple, static, individualised notions of research participation to a more complex, dynamic and inherently social account.
No preview · Article · Jan 2010 · Journal of medical ethics
[Show abstract][Hide abstract] ABSTRACT: This chapter examines the decisions of parents of critically ill babies who had not consented to take part in research. Their decisions were based on a misunderstanding of the information they had been given. It is suggested that patients may be overburdened with information and perhaps in this kind of case better decisions could be facilitated by providing less information rather than more.
[Show abstract][Hide abstract] ABSTRACT: The provision of bereavement care is an important part of neonatal intensive care. This systematic review of the effectiveness of interventions to support families and facilitate emotional adjustment following the death of a baby suggests that, while these are largely appreciated by parents who have participated in research, there has been little rigorous evaluation of their effectiveness. This review reflects on possible reasons for this; for example: NICU-led bereavement care is changing, the effectiveness of bereavement care is difficult to measure, concepts of effectiveness are not static, and ethical concerns complicate experimental research. Bereavement interventions are compassion-led and generally considered to be beneficial. New research questions and new methodological challenges are discussed with reference to two examples of evolving practice: bereavement photography and the use of ritual. Future research using innovative and sensitive RCTs and consensus amongst relevant stakeholders is suggested.
No preview · Article · Jun 2008 · Seminars in Fetal and Neonatal Medicine
[Show abstract][Hide abstract] ABSTRACT: The term 'therapeutic misconception' (TM) was introduced in 1982 to conceptualize how some psychiatry trial participants perceived and interpreted their involvement in research. TM has since been identified in many settings and is a major component in research ethics discussions. A qualitative study included a subgroup of interviews with five parents (two couples, one mother) who declined to enrol their baby in a neonatal trial. Analysis suggested the possibility of a counterpart to TM which, given the original terminology, we term the 'injurious misconception' (IM). While TM is closely linked to the elision of care and research, and involves an over-stated sense of benefit and protection, IM may be a product of a particularly keen and discomforting sense of distinctions between care and research and a correspondingly over-stated sense of risk and threat.
[Show abstract][Hide abstract] ABSTRACT: Objectives: To identify factors associated with good and poor recruitment to multicentre trials. Data sources: Part A: database of trials started in or after 1994 and were due to end before 2003 held by the Medical Research Council and Health Technology Assessment Programmes. Part B: interviews with people playing a wide range of roles within four trials that their funders identified as ‘exemplars’. Part C: a large multicentre trial (the CRASH trial) of treatment for head injury. Review methods: The study used a number of different perspectives (‘multiple lenses’), and three components. Part A: an epidemiological review of a cohort of trials. Part B: case studies of trials that appeared to have particularly interesting lessons for recruitment. Part C: a single, in-depth case study to examine the feasibility of applying a businessorientated analytical framework as a reference model in future trials. Results: In the 114 trials found in Part A, less than one-third recruited their original target within the time originally specified, and around one-third had extensions. Factors observed more often in trials that recruited successfully were: having a dedicated trial manager, being a cancer or drug trial, and having interventions only available inside the trial. The most commonly reported strategies to improve recruitment were newsletters and mailshots, but it was not possible to assess whether they were causally linked to changes in recruitment. The analyses in Part B suggested that successful trials were those addressing clinically important questions at a timely point. The investigators were held in high esteem by the interviewees, and the trials were firmly grounded in existing clinical practices, so that the trial processes were not alien to clinical collaborators, and the results could be easily applicable to future practice. The interviewees considered that the needs of patients were well served by participation in the trials. Clinical collaborators particularly appreciated clear delineation of roles, which released them from much of the workload associated with trial participation. There was a strong feeling from interviewees that they were proud to be part of a successful team. This pride fed into further success. Good groundwork and excellent communications across many levels of complex trial structures were considered to be extremely important, including training components for learning about trial interventions and processes, and team building. All four trials had faced recruitment problems, and extra insights into the working of trials were afforded by strategies invoked to address them. The process of the case study in Part C was able to draw attention to a body of research and practice in a different discipline (academic business studies). It generated a reference model derived from a combination of business theory and work within CRASH. This enabled identification of weaker managerial components within CRASH, and initiatives to strengthen them. Although it is not clear, even within CRASH, whether the initiatives that follow from developing and applying the model will be effective in increasing recruitment or other aspects of the success of the trial, the reference model could provide a template, with potential for those managing other trials to use or adapt it, especially at foundation stages. The model derived from this project could also be used as a diagnostic tool if trials have difficulties and hence as a basis for deciding what type of remedial action to take. It may also be useful for auditing the progress of trials, such as during external review. Conclusions: While not producing sufficiently definitive results to make strong recommendations, the work here suggests that future trials should consider the different needs at different phases in the life of trials, and place greater emphasis on ‘conduct’ (the process of actually doing trials). This implies learning lessons from successful trialists and trial managers, with better training for issues relating to trial conduct. The complexity of large trials means that unanticipated difficulties are highly likely at some time in every trial. Part B suggested that successful trials were those flexible and robust enough to adapt to unexpected issues. Arguably, the trialists should also expect agility from funders within a proactive approach to monitoring ongoing trials. Further research into different recruitment patterns (including ‘failures’) may help to clarify whether the patterns seen in the ‘exemplar’ trials differ or are similar. The reference model from Part C needs to be further considered in other similar and different trials to assess its robustness. These and other strategies aimed at increasing recruitment and making trials more successful need to be formally evaluated for their effectiveness in a range of trials. Not peer reviewed Publisher PDF
Full-text · Article · Dec 2007 · Health technology assessment (Winchester, England)