Article Text

Download PDFPDF
Prognostic disclosure and quality of life in palliative care: a systematic review
  1. Sanhapan Wattanapisit1,2,
  2. Richard Wagland1 and
  3. Katherine Hunt1
  1. 1 School of Health Sciences, University of Southampton, Southampton, UK
  2. 2 Palliative Care Unit, Thasala Hospital, Thasala, Nakhon Si Thammarat, Thailand
  1. Correspondence to Dr Sanhapan Wattanapisit, School of Health Sciences, University of Southampton, Southampton SO17 1BJ, UK; th.sanhapan{at}gmail.com

Abstract

Introduction Prognostic disclosure is an important component of communication in palliative care. Disclosing information on poor prognosis may affect quality of life (QoL) of palliative care patients. However, the effects of prognostic disclosure on QoL across different cultures and countries are unclear.

Objective To review the effects of prognostic disclosure on QoL of palliative care patients.

Methods A systematic review was conducted across seven databases (AMED, CINAHL plus, Cochrane Library, Medline (via the PubMed interface), Embase, Scopus and Web of Science). All primary studies, of any design, that explored the effects of prognostic disclosure on QoL of adult palliative care patients were eligible.

Results A total of 1926 records were screened for eligibility. Twenty-five articles were included (11 cross-sectional, 10 cohort, 3 mixed methods and 1 qualitative study). Studies were conducted in 11 countries. Five studies reported the sources of prognostic disclosure, while 20 studies did not. Emotional QoL was the most reported domain among the studies. The effects of prognostic disclosure on emotional aspects, overall QoL and other domains, including symptoms, physical functions, role functions, social functions and cognitive functions, were inconsistent.

Conclusions The effects of prognostic disclosure on QoL across cultures and countries are inconsistent. Cultural differences are not sufficient to explain the effects. Future research is needed to explore the association between prognostic disclosure and QoL, and develop tools to support clinicians to share prognostic information in the most sensitive and supportive way.

  • communication
  • end of life care
  • quality of life
  • prognosis

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Palliative care is provided to patients with life-threatening diseases and their families to maximise quality of life (QoL), and involves physical comfort, psychological well-being, spiritual well-being, social functioning and well-being, physical functioning and cognitive functioning.1 2 Communication is central to this care, and essential aspects include delivering bad news, discussing prognosis, developing goals of care and making a decision on treatment at the end of life.3 Good communication in palliative care means taking into account supporting emotions, sharing diagnosis and prognosis, and discussing treatment and advance care plans.3 Patient understanding of their disease status and prognosis is an important part of advance care planning and ensuring that they have an opportunity to express and meet their preferences for care at the end of life (ie, decision on cardiopulmonary resuscitation, life-prolonging treatment and place of death).4 5

Prognostic disclosure is fundamental to good end of life care5 and patients have the right of autonomy to be informed about their conditions. Accurate prognostic understanding promotes open treatment discussions between patients at the terminal stage and physicians.6 Communicating prognosis has been shown to help develop the goal of care, increase patient satisfaction, decrease hospitalisation and aggressive treatments, and enhance the use of hospice and palliative care services.7 However, in some cultures, disclosing prognosis is a challenging task for healthcare professionals due to cultural norms, expectations and beliefs. For example, disclosure may be considered inappropriate because it is bad luck or will discourage hope.8–10 This can make it difficult to provide palliative and end of life care services.

Physicians play an important role in discussing prognosis with patients. However, a study conducted in an Asian setting reported that only 56% of patients, who were aware of their prognosis, had been informed by their physicians, while 28% of them had worked it out for themselves, and 10% of them had been told by a family member.11 Other studies conducted in different cultures showed that between 30% and 78% of people living with advanced cancer were found to be accurately aware of their disease status.12–15 There is international variation on sharing and withholding prognosis. In some countries, family members can influence healthcare professionals’ communication to suit their wishes4 16 and to conceal prognostic information from patients.10 17 This situation is most common within cultures where the bond of family is firm and family members take a vital and central role in the care of patients.18

Previous studies conducted in China, Japan, South Korea and Greece have reported that families withheld patients’ information on poor prognoses because of concerns about a negative impact, such as, discouraging hope, distressing emotions and accelerating death.8–11 Indeed, emotional distress during the terminal stage has been found to affect a patient’s QoL19; the focus of palliative care.1 A cohort study from South Korea reported that patients with advanced cancer, who were aware of their terminal status, had a lower QoL than those who were not aware.13 In contrast, several studies have found that patients who were aware of their prognoses reported a better QoL than those who were unaware.14 20 Due to these conflicting findings, this systematic review aims to explore the effects of prognosis disclosure on QoL of palliative care patients.

Methods

Study design

A systematic review of published literature was conducted. The review included all primary study types and designs: quantitative, qualitative and mixed methods studies. The scope of the review followed the PICO mnemonic (population, intervention, comparison, and outcome).

The population was adult palliative care patients, defined within included studies as patients with terminal stage of illness that was not responsive to curative treatments. The intervention was prognostic disclosure, including any method to inform patients about their prognosis, or increase their awareness of their prognosis. The comparison included no comparator or any non-disclosure (eg, prognosis withholding, collusion between families and physicians). The outcome was any domain of QoL.

Eligibility criteria

Inclusion criteria

  • Primary studies in peer-review journals.

  • Studies that explored the effects of prognostic disclosure on QoL.

  • Studies that recruited adult patients (aged 18 years and above) with any diseases in the palliative or end-of-life stage.

Exclusion criteria

  • Conference abstracts, case reports, case series, letters to the editor, opinion papers.

  • Secondary studies or reviews (eg, systematic reviews, narrative reviews).

Search methods

Seven electronic databases, including AMED, CINAHL plus, Cochrane Library, Medline (via the PubMed interface), Embase, Scopus and Web of Science, were systematically searched from inception of the databases to 29 January 2019. The search strategy was developed and discussed by two authors (SW and KH). The first search strategy (Search A) was (palliative OR “end of life” OR terminal OR dying OR “approaching death”) AND prognos* AND (communicat* OR discuss* OR conversation OR disclos* OR reveal* OR deliver* OR withhold* OR collusion OR inform*) AND (“quality of life” OR QOL OR “health-related quality of life” OR HQOL) NOT (child* OR pediatric* OR paediatric*). Medical subject headings were also used (see ‘online supplemental file 1’). The records from Search A were initially scanned and few qualitative studies were found. Subsequently, the authors added the search terms for qualitative studies and developed another search strategy (Search B). Search B consisted of the ‘Search A’ and the additional search terms, AND (experience* OR narrative* OR perspect* OR qualitat* OR view*). Search B was conducted in each of the seven databases from inception of the databases to 8 March 2019. An updated search from March 2019 to 15 August 2020 was performed at the time of manuscript revision to recruit recent studies.

Supplemental material

Data collection and analysis

Study selection

All studies found in the databases were transferred to Endnote X4 citation manager (Thomson Reuters, Toronto, Ontario, Canada), and duplicates were excluded. Titles and abstracts were first screened by one author (SW), and subsequently discussed by two authors (SW and KH). Potentially eligible full-text articles were initially assessed by one author (SW) and then discussed and confirmed by another author (KH or RW). Disagreement between authors was resolved by discussion until a consensus was reached.

Data extraction

The data extraction form was designed to gather specific information for this review, including name of the first author, publication year, country, aims of the study, methods, participants, setting, prognostic disclosure assessment, key results about the effects on QoL. Two authors (SW and KH) piloted the data extraction form by extracting two studies separately and discussed the results to check agreement and appropriateness of the data extraction tool. One author (SW) extracted all the data. A second author (KH or RW) crosschecked the data extraction.

Quality assessment

Eligible articles were assessed for methodological quality using the Mixed Methods Appraisal Tool (MMAT) version 2011.21 The tool is suitable and reliable for systematic reviews of studies of diverse methodologies.21 There are four criteria for each study type. Overall scores were calculated when each study met 1–4 criteria. The score was presented as 25%–100%. For qualitative and quantitative studies, the scores ranged from 25% (one criterion met) to 100% (four criteria met). The scores of mixed methods studies were calculated by the combination of qualitative, quantitative and mixed methods criteria. The overall score of a mixed methods study could not exceed the score of the weakest criteria (criteria for qualitative, quantitative or mixed methods study).

Data analysis and synthesis

One author (SW) analysed and synthesised all the data from the data extraction forms. Two authors (KH and RW) verified the analyses. The data analysis and synthesis were performed based on the characteristics of the data of each included study (ie, qualitative or quantitative data). The final results were confirmed by discussion and consensus within the team.

Subgroup analysis

Based on the initial analysis, the authors found that a proportion of the studies reported the sources (ie, physicians, family members) of prognostic disclosure. The majority of studies described prognostic awareness or understanding without information about sources and methods of prognostic disclosure. Therefore, the included studies were divided into two categories: (1) prognostic disclosure studies and (2) prognostic awareness studies. The authors analysed the effects of prognostic disclosure or awareness on QoL domains separately.

Results

Search results

The search results consisted of 1926 records after removal of duplicates. A total of 40 full-text articles were assessed for eligibility. Subsequently, 15 articles were excluded and 25 articles were eligible and included in the synthesis (figure 1).

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram.

Quality assessment

The quality scores of included studies ranged from 50% to 100%. Nine studies scored 50%,14 22–29 11 studies scored 75%11 15 30–38 and 5 studies scored 100% (online supplemental file 2).12 13 20 39 40 The detail of the quality assessment using the MMAT is presented in ‘online supplemental file 2’.

Supplemental material

Study characteristics

The included articles consisted of 11 cross-sectional studies,12 15 20 22 23 25 27 30 32 34 36 10 cohort studies,11 13 14 26 28 29 35 37 38 40 3 mixed methods studies24 31 33 and 1 qualitative study.39 The included articles derived from 11 countries including, the USA (n=7),15 25 28 31 32 35 40 South Korea (n=6),11 13 14 34 37 38 Taiwan (n=3),26 27 29 Australia (n=1),33 Austria (n=1),24 Cuba (n=1),23 Belgium (n=1),22 Turkey (n=1),36 Hong Kong (n=1),30 India (n=1)20 and Japan (n=1)12 (online supplemental file 3). One study was conducted in two countries, USA and Australia.39 Participants of all studies were patients with advanced stage cancers. The mean age of participants ranged from 50 to 74 years old. The sample size of the studies varied between 34 and 935.

Supplemental material

Of 25 studies, the ‘prognostic disclosure studies’ consisted of five studies that revealed the sources of prognostic disclosure.11 20 24 33 40 Three of these studies reported prognostic disclosure was made by physicians.24 33 40 A study by Rumpold et al described that physicians would inform the prognosis based on patients’ request.24 One cohort study asked whether the information was disclosed by physicians or families, or whether the patient had perceived it themselves.11 One study compared the effects of prognostic disclosure between patients who were disclosed the prognosis; and patients who were concealed the prognosis by a collusion between families and physicians.20 These five studies did not report the processes, techniques and contents of prognostic disclosure.

The ‘prognostic awareness studies’ were the remaining studies (n=20) that did not identify the sources of prognostic disclosure. These studies explained patients’ awareness or understanding of disease status. The processes to assess patients’ awareness included simple questions,12 14 23 26 27 29 34–36 40 questionnaires15 25 28 30 32 37 or interviews.13 31 38 39

Effects of prognostic disclosure on QoL

QoL assessment

Different QoL assessment methods were applied in the five prognostic disclosure studies (online supplemental file 3). Two studies used the European Organization for Research and Treatment of Cancer—Core Quality of Life Questionnaire (EORTC QLQ-C30) to assess QoL.11 20 One study assessed QoL domains by using a combination of several tools, including McGill Quality of Life Index’s, Structured Clinical Interview for Diagnostic and Statistical Manual-Fourth Edition and Endicott Scale for major depressive disorder.40 Two studies assessed emotional domain by using different methods. A study of Rumpold et al assessed patients' emotional response in six dimensions,24 while Gattellari et al highlighted the anxiety level by using 20-item Spielberger State-Anxiety Form.33

Overall QoL

Two studies presented the effects of prognostic disclosure on overall QoL. Both studies reported a positive impact of prognostic disclosure on overall QoL (table 1).11 20 A study by Victor et al found that patients, who were disclosed the prognosis by physicians, had a higher score of overall QoL compared with those who were concealed the prognosis.20 Yun et al reported a positive impact on overall QoL among patients who were informed their prognoses by physicians or families.11

Table 1

The effects of prognostic disclosure and prognostic awareness on quality of life domains

Symptoms

Three studies reported the effects of prognostic disclosure on patients’ symptoms (table 1). The results were uncertain. Two studies showed a positive impact on reduction in pain and fatigue.11 20 A study reported more symptom burden which was a negative impact of prognostic disclosure.40

Emotions

All five studies presented the effects of prognostic disclosure on emotional QoL domain (table 1). Three studies reported no impact.20 33 40 One study reported a positive impact: the prognostic disclosure group had a higher emotional score than the non-disclosure group (self-perceived).11 In contrast, one study showed significantly higher emotional distress among patients who received prognostic disclosure by physicians.24

Physical functions

Physical functions involved the ability to perform activities such as walking, being restricted to a chair or bed, feeling ill or lack of energy. Three prognostic disclosure studies focused on the effects on physical functioning (table 1). The results of two studies reflected a higher score of physical functioning, which was a positive impact on prognostic disclosure.11 20 Another study reported no effect on physical functions.40

Role functions

Two studies reported role functioning scores by using the same questionnaire (EORTC QLQ-C30). The two studies presented different results. A study by Yun et al reported a positive impact,11 while Victor et al reported no impact (table 1).20

Social functions

Three out of five prognostic disclosure studies reported social functions as the outcomes (table 1). One study reported a positive association between prognostic disclosure and social functions.11 Two studies showed no association.20 40

Cognitive function

Two studies evaluated cognitive function (table 1). Both studies reported that patients who had been told their illness was terminal had a higher score on the EORTC QLQ-C30 Cognitive Functioning Scale than patients who had not been told.11 20

Effects of prognostic awareness on QoL

QoL assessment

Among 20 prognostic awareness studies, all quantitative studies used validated quality assessment tools or a modified version of the tools (online supplemental file 3). Examples of validated tools used included the EORTC QLQ-C30,14 34 36 Functional Assessment of Cancer Therapy (FACT)12 15 31 32 and McGill QoL Questionnaire.26–29 35 37 40 These tools evaluated QoL in overall and specific domains including symptoms; physical functioning; emotional, role and social functioning; and cognitive functioning. One qualitative study evaluated QoL from semistructured interviews and a focus group process.39

Some tools evaluated specific QoL domains. The Hospital Anxiety and Depression Scale was used to evaluate emotions,14 15 22 25 26 29 31 and the 10-item Self-Perceived Burden Scale (SPBS) were used to evaluate spiritual suffering.26 table 1 identifies the effects on QoL in each domain, that is, overall QoL, symptoms, emotions, physical functions, role functions, social functions, cognitive functions and spiritual or existential well-being.

Overall QoL

Fourteen studies explored the effects of prognostic disclosure on general QoL (table 1). The overall QoL emerged from the total score of subscales12 15 26 27 31 32 38 or questions asking the global QoL status.13 14 22 34 36 37 39 The result on this domain was inconsistent. Six studies reported a negative impact of prognostic awareness on overall QoL.12 13 15 26 32 37 In contrast, a positive effect of terminal prognostic awareness on overall QoL was found in two studies.27 39 One qualitative study showed prognostic disclosure promoted patients to accept ways to enjoy a good QoL.39 Another six studies reported no effects.14 22 31 34 36 38

Symptoms

Six studies presented the effects of prognostic awareness on the severity of patients’ symptoms (table 1). Overall, results were mixed. Two studies showed that patients who understood the prognosis reported lower symptoms, including appetite loss, fatigue and constipation.14 23 Two studies described no effects of awareness on symptoms.34 38 Two studies reported a negative impact on one symptom; increased pain.22 37

Emotions

The majority of included studies presented emotional outcomes. Examples of emotional outcomes included anxiety, depression or nervousness. Sixteen studies reported the effects of prognostic awareness on emotion (table 1), the results of which were ambiguous. Three studies showed positive outcomes on emotions,14 23 30 especially, reduced anxiety.14 23 30 Five studies described no impact13 22 31 34 38 while seven studies showed negative impacts on emotions,12 15 25 26 29 32 37 four of which reported increased depression.15 25 26 29 A study showed a fluctuation in psychological well-being over time among patients who perceived more accurate in prognostic understanding.28

Physical functions

Nine studies presented the different outcomes on physical functioning and well-being (table 1). Six studies presented no association between prognosis awareness and physical functions.12 14 22 29 32 34 One study reported a higher physical functioning subscale score among patients who had been told that their condition was terminal,38 while two studies reported patients who aware of their terminal prognostic status perceived worse physical well-being.35 37

Role functions

Role functions were assessed using the role functioning subscale from the EORTC QLQ-C30, the 36-item Short Form Health Survey and functional well-being subscales from the FACT, which assessed the ability to work, engage in hobbies or enjoy everyday activities. Role functions were reported in five studies (table 1). These results were also conflicting. Two studies showed positive outcomes, reporting that patients who were aware the terminal status or aware that their condition was incurable had higher role functioning than patients who had not been told or were not aware of their disease status.14 38 A negative outcome was reported in one study, with poorer functional well-being reported among patients who accurately perceived their terminal prognosis than patients who inaccurately perceived their terminal prognosis.12 Two studies showed no difference.32 34

Social functions

Eight studies described the effects on social functions, which were the interactions between patients and partners, families or friends (table 1). The results were variable. A positive effect on social functions was reported in three studies.14 23 38 In one study, patients who were aware of their terminal status had a significantly higher social functioning score from the EORTC QLQ-C30 than those who were not aware.14 A study by Justo Roll et al showed patients who were aware of terminal illness were less likely to report lack of support than patients who were unaware.23 Three studies showed negative social outcomes.26 32 37 One of these three studies measured self-perceived sense of burden to others and found that this was higher in patients who had an accurate awareness of their terminal prognosis.26 Another study showed patients who understood their cancer was incurable were more likely to report low FACT-G social well-being score than patients who thought that the cancer was likely to be cured.32 Two studies showed no association.12 34

Cognitive function

Four studies presented the influence of prognostic awareness on cognitive function. Similar to the other QoL dimensions, the impact of prognostic awareness on cognitive function varied between studies (table 1). The effects were reported by four studies, with findings varying from positive,38 negative35 and no effect.14 34 One study showed a positive outcome measured by the Mini-Mental Status Examination.38 Conversely, a study that measured cognitive function using the FACT-Cognition found that poorer scores on cognitive function were associated with awareness of illness being terminal.35

Spiritual or existential well-being

Two studies reported on spiritual or existential well-being (table 1). A study assessed spiritual or existential suffering by using the 10-item SPBS. It was found that participants who accurately understood their terminal status significantly reported high self-perceived sense of burden.26 Another study presented that patients in the awareness group had lower existential well-being scores.37

Effects of prognostic disclosure and prognostic awareness on QoL by country

Table 2 shows the results of the effects of prognostic disclosure and prognostic awareness on the overall QoL and the top three reported QoL domains among the included studies by country. The most reported domains were emotions (n=21), physical functions and well-being (n=12), and social functions (n=11). The five prognostic disclosure studies were conducted in five different countries. According to the prognostic awareness studies, a set of studies (more than one study in each country) were analysed for studies conducted in the USA, Taiwan and South Korea. For overall QoL, the findings within each country were mixed (a combination of positive outcomes, negative outcomes and no effect). The results of the other three domains were also mixed. However, two studies from Taiwan illustrated negative results of the emotional domain.26 29

Table 2

Effects of prognostic disclosure and prognostic awareness on overall, emotions, physical functions and well-being, and social functions categorised by countries

Discussion

This systematic review sought to identify the effects of prognostic disclosure on QoL domains in palliative care patients. A total of 25 studies were included from 11 countries in several regions. Five studies identified the sources of prognostic disclosure. Two studies directly reported on the impact of prognostic disclosure on overall QoL and cognitive domain, and both of these reported that prognostic disclosure had a positive impact of on overall and cognitive QoL. The results of other domains, including symptoms, emotions, physical functions, role functions and social functions, were inconsistent.

Twenty studies discussed prognostic awareness but did not identify the sources of prognostic disclosure. The majority of the studies investigated the relationships between prognostic awareness and emotional aspects. The effects of prognostic awareness on QoL varied between studies. The findings of most QoL domains, including overall; symptoms; emotional aspects; physical functions; role functions; social functions; and cognitive functions, varied between studies. Two studies reported a negative impact of prognostic awareness on spiritual or existential well-being. Even when studies of the impact of prognostic awareness on QoL had been conducted in the same country, they still reported conflicting findings, with evidence of positive outcomes, negative outcomes and no effect.

This systematic review has showed that prognostic disclosure may lead to positive outcomes of overall QoL and cognitive domain. However, the findings should be interpreted carefully because there were only two studies (and they both used the same outcome measure) that investigated the effects of prognostic disclosure on overall QoL and cognitive domain. Although they did not directly address whether prognostic disclosure had an impact on QoL, the prognostic awareness studies included in this systematic review could not be excluded because patients in those studies may have been informed of their prognosis by physicians and families, or they may have guessed themselves. For this reason, both ‘prognostic disclosure’ and ‘prognostic awareness’ studies formed the data analysis and synthesis of this systematic review.

The findings from the prognostic awareness studies demonstrated unpredictable QoL outcomes and these may affect QoL differently over time. There was no pattern of results between different cultures. In other words, the cultural backgrounds of patients should not be an indicator to decide whether or not to disclose prognosis. However, concepts of death and dying do vary across cultures, but there is no evidence to suggest that these differences impact on decision-making and preferences. A study conducted in three countries (USA, Germany and Israel) revealed that culture, nationality and religion were not predictors of end of life decision-making and attitudes towards death (ie, advance directives, life-sustaining treatment termination, legal frameworks regarding end of life care).41 This supported the inconsistent findings of this systematic review. Rather, the effects of prognostic disclosure on QoL are more likely to be influenced by communication techniques, the nature of the information provided on disease status, the patient’s personality or characteristics and individual coping strategies.42–44

Most studies emphasised overall QoL and emotional aspects. Indeed, emotional distress, such as anxiety and depression, was a common issue among patients with advanced diseases.45 46 Moreover, evidence showed that psychological disorders affected physical symptoms and patients’ social and existential concerns.47 In contrast, only two studies highlighted spiritual or existential well-being.26 37 This might be because this systematic review focused on comprehensive QoL rather than spiritual or existential well-being. Another reason was spiritual or existential aspects might be considered as another specific domain, and not a dimension of QoL.48

Some studies showed that disclosing and perceiving prognosis could lead to negative outcomes. This is an important finding for practice and highlights that clinicians should be aware of their patients’ personal characteristics when discussing prognosis, such as their personality, coping mechanisms, readiness to receive such information and information preferences.49 50 Most importantly, appropriate communication techniques could be a key factor to prevent negative QoL outcomes, which is the main goal of palliative care.4 51 Various communication techniques were recommended. For example, delivering information in a private setting, exploring patients’ thoughts before sharing the information, clarifying medical terms, communicating with compassion, conveying honesty, providing trust and hope, allocating enough time for the discussion, avoiding disruption, rechecking the patient’s understanding and providing an opportunity for questions.42 52

Other possible approaches to prevent negative outcomes are building a good doctor–patient relationship and promoting continuity of care. Patients can experience discomfort when dealing with new doctors, and can find it exhausting to recite the story of their illness and experience from the beginning each time they meet a new healthcare professional.52 Patients appreciate seeing the same doctors and nurses with whom they can build a rapport, who are open to communication and who treat them with respect.51 52 Good continuity of care can help clinicians and patients to connect with each other as well as facilitate the sharing of information about disease status, prognosis and treatments.

Rather than a model of care that excludes patients or alienates family, discussions about prognosis can involve caregivers, and information can be individualised to their level of understanding and willingness to discuss with clinicians.53 Although we cannot say for certain based on what is reported in the included studies, the way that bad news is delivered is likely to be related to QoL outcomes. Training or continuing professional development to improve communication skills should be available to healthcare professionals to increase their competence and confidence in discussing prognosis with their patients.

There is a need to consider and implement a practical tool for prognostic disclosure in palliative care. Developing such a tool will require an understanding of the differences of personal preferences and could support clinicians in the evaluation process prior to prognostic disclosure, to decide on communication approaches, and the supportive care after disclosing the information. Studies have suggested the use of question prompt lists or patient decision aids to support prognostic discussion and treatment decision-making at the end of life.54 55

Based on the variable impact of prognostic disclosure on QoL, future research should focus on two areas. First, there is a need to explore the factors that affect QoL after the process of prognostic disclosure. Understanding the factors related to QoL could help healthcare professionals to preempt and then deal with a patient’s reactions. Second, future research should focus on how to conduct prognostic disclosure in the most sensitive way, to reduce the negative effects on QoL.

This systematic review had several strengths. First, the search strategy covered the major databases in the field of health. The authors performed the additional search (Search B), to include more qualitative studies, after the initial screening. This led to a higher yield of search results. In addition, an updated search was performed at the time of manuscript revision to gather recent published studies. Second, the review included studies from different countries and regions. This approach could explain the effects of prognostic disclosure on QoL in cross-cultural settings. Third, this systematic review consisted of the subgroup analyses of both prognostic disclosure studies that identified the sources of prognostic information; and prognostic awareness studies which did not provide information about the sources of prognostic disclosure. Lastly, this review performed data synthesis and analysis of various QoL domains.

There were three limitations of this systematic review. First, participants were all patients with advanced stage cancers. This may not represent people who had other advanced diseases. Second, the systematic review could not identify methods used to disclose prognoses among the included studies. Communication methods could be a confounding factor, which had an impact on QoL scores. Third, we are unable to determine the causality of the results. For example, negative outcomes of QoL might be an effect of prognostic disclosure or an effect of the deterioration in conditions of the patients.

Conclusions

Patients have a right to prognostic information, and the disclosure of this information is an important communication issue in palliative care. This systematic review presents variable and inconsistent effects of prognostic disclosure on overall and specific QoL domains, and these are not dependent on country or culture. Future research should focus on factors that affect QoL after the process of prognostic disclosure, intercultural aspects of palliative care and development of a prognostic disclosure tool (eg, patient decision aids), that can minimise the negative effects on QoL.

Ethics statements

Patient consent for publication

Acknowledgments

The authors would like to acknowledge Chevening Scholarships, the UK government’s global scholarship programme, funded by the Foreign and Commonwealth Office (FCO) and partner organisation for granting SW to study at the University of Southampton. The authors also would like to thank the librarians at the University of Southampton Academic Skills Hub for their help in developing the search strategy, and Apichai Wattanapisit for his support.

References

Supplementary materials

Footnotes

  • Contributors SW and KH were responsible for concept or design of the work. All authors were responsible for the acquisition, analysis and interpretation of the data. SW wrote the first draft of the manuscript. KH and RW revised the draft. All authors approved the final version of the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.