Unmet supportive care needs associated with quality of life for people with lung cancer: A systematic review of the evidence 2007–2020

1 INTRODUCTION

Lung cancer is the most common malignancy and the most common cause of cancer death worldwide (Wong et al., 2017), with an estimated 2.1 million new cases and 1.8 million deaths in 2018 (World Health Organisation, 2018). In spite of several advancements in the treatment protocols, international 5-year survival rates remain lower than other causes of cancer at around 17% (Wong et al., 2017). Treatment options for people with lung cancer are determined by a number of factors including histological findings, stage of cancer at diagnosis, patients' performance status and the patients' preferences (Spiro & Hardavella, 2013). Patients, therefore, may undergo a range of complex treatments, including surgery, chemotherapy and/or radiotherapy depending on the cancer stage at the time of diagnosis (Detterbeck et al., 2013). Recently, targeted therapies for patients with specific molecular mutations have led to some improvements in survival (Al Achkar et al., 2020). Nevertheless, many patients are diagnosed with lung cancer at an advanced stage, partly because the early stages of the disease are often asymptomatic (Pozo et al., 2014), symptoms can be misinterpreted leading to delays in referral and diagnosis (Walter et al., 2015) and there is a lack of awareness of symptoms and negative attitudes towards the disease (Global Lung Cancer Coalition, 2018). Whichever factors are involved, late presentation can limit the provision of treatment with curative intent (Maguire et al., 2013).

As a result of many of the issues described above, the symptom burden experienced by patients with lung cancer has been reported as greater than that experienced with other types of cancer (Sung et al., 2017). Symptoms associated with the range of treatments and the disease itself, include pain, dyspnoea, fatigue and anorexia, and are associated with impaired functioning and have a negative impact on quality of life (QoL) (Iyer et al., 2014). Individuals who receive a lung cancer diagnosis also experience high rates of anxiety and depression (Haun et al., 2014) and are at a greater risk for suicide compared with other cancers (Anguiano et al., 2012). This may, in part, be related to the side-effects of aggressive treatment, poor prognosis and the stigma associated with smoking behaviours (Liao et al., 2011). Indeed, the perceived stigma and shame arising from the social perception that lung cancer patients are responsible for their illness (Chapple et al., 2004) could lead to concealment of the lung cancer diagnosis (Gonzalez et al., 2015). Furthermore, the stigma and high levels of depression associated with lung cancer are likely to further reduce the QoL of patients (Brown et al., 2014).

The complex range of needs associated with lung cancer has emphasised the importance of supportive care to improve the QoL of patients regardless of prognosis (Molassiotis et al., 2015). While there are variations in the definitions of supportive care, all of them include elements of symptom management and aim to improve QoL (Hui, 2014). Thus, a supportive care perspective could be viewed as one which promotes and sustains well-being, while enhancing the patient's capacity to cope with the demands of diagnosis and treatment (Molassiotis et al., 2015). For the purposes of the current study, we have defined supportive care as: ‘The provision of the necessary services for those living with or affected by cancer to meet their physical, emotional, social, psychological, informational, spiritual and practical needs during the diagnostic, treatment, and follow-up phases, encompassing issues of survivorship, palliative care and bereavement’ (Fitch, 2008, p. 11).

A systematic review of the supportive care needs of people diagnosed with lung cancer identified a broad range of needs associated with the condition (Maguire et al., 2013). While some of the needs are relevant to many types of cancer, others may be more pervasive in lung cancer. These include needs related to coping with the physical effects of the disease and the demands of treatment (Li & Girgis, 2006), as well as the need for appropriate supports to manage the high levels of psychological distress (Brown et al., 2014). Critically, there is evidence that for many patients these needs are largely unmet, and there is some indication that lung cancer patients have a higher rate of unmet care needs relative to other cancer patients (Li & Girgis, 2006; Liao et al., 2011).

Although supportive care needs of lung cancer survivors have been systematically examined in the literature, the relationship between these needs and QoL is unclear in the literature to date. This relationship is important because identifying which unmet needs have the greatest implications for the QoL would be useful in developing and prioritising interventions and services to address these needs. While the association between unmet needs and QoL has been assessed in other patient groups (e.g., breast cancer: Aranda et al., 2005; e.g., dementia: Miranda-Castillo et al., 2010), it remains important to understand the relationship specific to lung cancer. The often protracted course of the disease, the high symptom burden and stigma associated with the disease all suggest that lung cancer may have a differing relationship between unmet needs and QoL compared with other disease groups (Li & Girgis, 2006; Liao et al., 2011; Maguire et al., 2013). A number of studies have indicated that QoL is closely linked to symptom burden and severity in lung cancer (Anant et al., 2005; Shi et al., 2016), yet to date, less is known about the relationship between QoL and broader unmet needs, such as psychological, informational and practical care needs (Maguire et al., 2013). Furthermore, as one of the aims of supportive care is to improve and maintain QoL, it is important to understand the impact that such unmet needs may have on the QoL of patients. Therefore, the aim of this systematic review was to examine the current evidence on the relationship between QoL and unmet needs of patients with lung cancer during treatment and recovery.

2 METHODS

This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and has been registered with Prospero (CRD42019126861).

2.1 Search strategy

A systematic search of six databases (MEDLINE, CINAHL, EMBASE, PsychINFO, PubMed and Web of Science Social Sciences Citation Index) was conducted by the first author for the period between January 2007 and March 2020. The search was limited to literature published in English and further limited to publications from the last 12 years to reflect current practice. See Appendix 1 for the MEDLINE search strategy. The first author also screened the reference lists of included articles and existing reviews to identify any additional eligible articles. All titles and abstracts of the identified articles were screened for eligibility by two authors (AC and SW), and those that were deemed irrelevant were discarded. The full texts of the remaining studies were obtained and independently reviewed by the same two authors. Any disagreement as to inclusion/exclusion was resolved by discussion and consensus, and the other authors (SD or PG) were available to resolve any disputes.

2.2 Eligibility criteria

Papers were eligible for inclusion if they were cross-sectional, longitudinal and intervention studies and if participants were aged 18 or over with a lung cancer diagnosis (small cell and nonsmall cell lung cancer). We also considered studies with mixed cancer populations if they otherwise met the inclusion criteria. In this case, study authors were contacted to determine if the lung cancer findings could be disaggregated. Finally, studies were included if they had investigated the relationship between QoL and patient/carer needs using a standardised measure of health-related QoL (either generic, e.g., SF-32 or cancer specific, e.g., European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire [EORTC QLQ]) and an explicit measure of supportive care needs (e.g., Concerns Checklist, Needs Evaluation Questionnaire and Supportive Care Needs Survey [SCNS]). With regard to exclusion criteria, studies focussing exclusively on patients receiving end-of-life care were excluded. Studies that reported QoL and supportive care needs but did not statistically analyse the relationship were also excluded.

2.3 Quality assessment

Included articles were critically appraised by two authors (AC and SW) to evaluate risk of bias using a 12-item checklist based on previous research (Dunne et al., 2017), supplemented by standard critical appraisal questions (Crombie, 1996). The studies were given a score of two for every criterion on the checklist which they fulfilled, a score of one where they partially fulfilled the criterion and zero where they failed to fulfil the criterion. We considered studies to be of high quality if they scored 17 or more out of a possible score of 24, adequate quality if they scored between 9 and 16 and low quality if they scored less than 9. If there was any disagreement between the raters, consensus was achieved through discussion.

2.4 Data extraction and synthesis of results

The first author extracted the following data from eligible articles: (1) study location, (2) aim(s), (3) design, (4) recruitment setting, (5) patient characteristics (age, gender, type of lung cancer, stage and treatment) and caregiver characteristics (age, gender), (6) QoL measure(s) used, (7) need measure(s) used and (8) whether or not QoL was associated with need and if so, the specific results in the overall population and any subgroups.

Due to the heterogeneity of the included studies, we were unable to conduct a meta-analyses and instead adopted a narrative approach to synthesise findings.

3 RESULTS

Six studies were included in this review. Figure 1 summarises the screening and selection process; searches of the electronic databases yielded 6241 studies. After removal of duplicates, we screened titles and abstracts of 3236 and identified 61 as potentially relevant. Following assessment of the full text of these potentially relevant articles, we identified four studies involving patients with lung cancer that met our inclusion criteria (Chan et al., 2012; Giuliani et al., 2016; Pardon et al., 2011; Sanders et al., 2010). A further six studies were identified with mixed cancer populations; the lung cancer cohort from one of these was reported in an included study (Chan et al., 2012). We contacted the authors of the remaining papers requesting the data for the lung cancer samples. We were provided with the relevant raw data for two of these studies (Molassiotis et al., 2017; Zhu et al., 2018). A subset of Molassiotis et al. was reported in a separate paper (Li et al., 2019), and thus, we have not included Li et al. in this review.

image

Prisma flow diagram

3.1 Study characteristics

All of the included studies were cross-sectional in design; no intervention or longitudinal study met our inclusion criteria. One study pooled data from several Asia-Pacific countries, two studies were conducted in China, two in North America and one in Europe. Study characteristics are presented in Table 1. The sample sizes ranged from 46 to 142. The combined total was 562 patients, with a mean age of 63.42 years (SD = 6.02; derived from five studies; one study reported age in categories) and 35% were female. Participants were recruited via hospital settings; the majority (n = 254; 60%) where reported, were diagnosed with Stage III or IV lung cancer. The time since diagnosis was reported in five studies, with most participants diagnosed less than 2 years (Table 1).

TABLE 1. Characteristics of included studies Author (year) Country Sample Clinical characteristics Measurement time-points QoL measure(s) Needs measure(s) Overall finding Chan et al. (2012) China

N = 48

Mean age: 61

Female: 25%

≤Stage II n = 13

≥Stage III n = 34

Treatment completed within the previous year.

Median time since diagnosis 12 months (IGR 10–18)

FACT-G

Total and subscales

SCNS-SF34

Unmet psychological and physical needs negatively associated with QoL

Giuliani et al. (2016)

Canada

N = 89

Median age: 71

Female: 51%

Mixed histology

Adenocarcinoma 49%

Squamous cell 16% ≤ Stage II n = 7 ≥ Stage III n = 14

Completed or undergoing treatment.

Time since diagnosis not reported

Overall health and overall QoL

(two items from EORTC-QLQ-C30)

SCNS-SF34

Lower QoL associated with increased levels of unmet needs across the five domains of the SCNS Molassiotis et al. (2017)a Asia- Pacific (42% China; 21% Thailand)

N = 142

Mean age: 59.59

Female: 33.1%

Not reported

Completed first line treatment.

Mean time since diagnosis: 46.27 months (range 6–318 months)

Single item CaSUN Lower QoL associated with greater needs across the 5 domains of the CaSUN

Pardon et al. (2011)

Belgium

N = 128 age: 48% ≤ 64 35% 65–74 17% ≥ 75

Female: 20%

Advanced NSCLC

Stage IIIb or IV

Newly diagnosed

Overall QoL

(one item from EORTC-QLQ-C15-PAL

Information preferences (in general and six-disease specific items) QoL higher when information needs related to chances of a cure were met Sanders et al. (2010) USA

N = 109

Mean age: 68.5

Female: 53.2%

NSCLC or small cell

≤Stage II n = 27 ≥ Stage III n = 49 Unknown = 33

Time since diagnosis 54.1 weeks (SD 63.6) Physical functioning subscale SF-36 SCNS-SF31 Lower physical functioning QoL associated with unmet needs across the four domains of the SCNS

Zhu et al. (2018)a

China

N = 46

Mean age: 57

Female: 29%

≤Stage II n = 16 ≥ Stage III n = 29

Receiving treatment.

Mean time since diagnosis 11.7 months (range 0–58 months)

EORTC-QLQ-C30

Global health and subscales

SCNS-SF34

Worse QoL associated with unmet needs on the physical domain of the SCNS Abbreviations: CaSUN, Cancer Survivors Survey of Needs; EORTC-QLQ, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire; FACT-G; Functional Assessment of Cancer Therapy-General; NSCLC, nonsmall cell lung cancer; QoL, quality of life; QLQ-C15-Pal, Palliative; SCNS-SF, Supportive Care Needs Survey- Short Form. a Disaggregated data provided by study authors. 3.2 Quality assessment

Two studies were rated as ‘high quality’ (Molassiotis et al., 2017; Zhu et al., 2018), while the remaining four were deemed as ‘adequate quality’. Areas in which studies were downgraded included failure to justify sample size and limited descriptions of recruitment strategies and nonresponders. Further details are available in Appendix 2.

3.3 Measures used

There was some variation across the studies in the tools used to assess QoL and unmet needs (Table 1). Four studies of the six studies assessed patients' QoL using cancer specific tools (Chan et al., 2012; Giuliani et al., 2016; Sanders et al., 2010; Zhu et al., 2018). Molassiotis et al., used a single item, taken from the Cancer Survivors Survey of Need (Filsinger et al., 2011); in all cases a higher score represented a higher QoL. Sanders et al (2010) employed the physical funcitioning scale of the SF 36 but did not report the scores. Further details of the QoL findings can be found in the supporting information.

In relation to measuring unmet needs, four studies (Chan et al., 2012; Giuliani, 2016; Sanders et al., 2010; Zhu et al., 2018) used the SCNS (Boyes et al., 2009). Three of the studie used the 34-item version which assesses the perceived need for help amonsgt people with cancer across five domains: (1) psychological, (2) health system and information, (3) physical and daily living, (4) patient care and support and (5) sexuality. Sanders et al. (2010) used a version of the SCNS-SF31 that does not include the sexuality domain (SCNS-SF31).

Three studies reported the mean total scores for unmet needs (Table 2); the scores indicated that the greatest need was within the health systems and information domain for two of the three studies (Chan et al., 2012; Zhu et al., 2018). Three studies (Chan et al., 2012; Giuliani et al., 2016; Sanders et al., 2010) reported that more than half of the participants had at least one unmet need on two or more domains (Table 2). The most frequently reported unmet needs across these three studies were the need to be informed about test results (58%; Chan, et al. 2012), fear of cancer spreading (52%; Giuliani et al., 2016) and lack of energy or tiredness (75%; Sanders et al., 2010)

TABLE 2. Mean total scores for unmet needs, and proportion (%) with at least one unmet need for each domain of the SCNS-SF Chan et al. (2012) Giuliani et al. (2016) Sanders et al. (2010) Zhu et al. (2018) Health systems and information 27.5 72.5% 20.1 52% - 66% - 38.6 Psychological 14.1 33.1% 21.1 66% - 78% - 30.5 Physical 20.5 41.7% 10.3 58% - 89% - 26.9 Patient care 14.2 50% 8.3 24% - 57% - 33.8 Sexuality 2.6 0% 4 20% Not measured - 12.1 Note: Means not reported in Sanders et al. (2010); % not reported in Zhu et al. (2018).

Molassiotis et al. (2017) used the Cancer Survivors Unmet Needs Survey (CaSUN; Hodgkinson et al., 2007). This measure includes 35 items and assesses unmet needs across five domains: (1) Existential Survivorship; (2) Comprehensive cancer care; (3) Information; (4) QoL and (5) Relationships. The mean score of unmet need was highest in the Existential survivorship domain (mean: 17.89) and lowest in Relationships (mean: 2.79). The most common unmet needs were related to being provided with up to date information and ‘fear of cancer coming back’ (44% and 42%, respectively, indicating proportion with moderate or strong unmet need); the other most endorsed needs are related to unmet needs in the Comprehensive Cancer Care domain (46% accessible local healthcare services, best medical care 42% and the need for doctors to talk to each other 41%).

Pardon et al. (2011) examined information needs only and assessed these needs and preferences using items adapted from other questionnaires (Kutner et al., 1999; Wen & Gustafson, 2004). The findings indicated that over 88% of patients had information needs related to their diagnosis, treatment options, chances of a cure and life expectancy; over half had information needs related to palliative care (63.5%) and end-of-life decisions (56.8%).

3.4 Association between QoL and unmet needs

Five studies reported a negative association between QoL and unmet needs. Higher levels of supportive care needs across the five domains of the SCNS were associated with lower overall QoL using a single item from the EORTC QLQ-C30 in one study (Giuliani et al., 2016), and four domains (i.e., study excluded the sexuality domain) were associated with poorer physical QoL using the physical functioning subscale of the SF-36 (Sanders et al., 2010), although the relationship in the former study was only found in the univariate but not the multivariable analysis. In one study (Chan et al., 2012), unmet physical and psychological needs were independently associated with lower QoL measured by the Functional Assessment of Cancer Therapy-General (FACT-G). In the study based in China (Zhu et al., 2018), only unmet psychological needs were negatively correlated with global QoL using the EORTC QLQ-C30.

Higher levels of unmet needs across the five domains of the CaSUN were also negatively associated with lower global QoL (Molassiotis et al., 2017).

Pardon et al. (2011) reported that patients who had their information needs met about chances of a cure had a significantly higher QoL using the EORTC QLQ-C15-PAL to rate overall QoL, than those who were not informed, but no relationship was observed between QoL and other information needs and preferences.

4 DISCUSSION

This systematic review is the first, as far as we are aware, to provide preliminary evidence that a range of unmet supportive care needs is associated with poorer QoL for people with lung cancer during and after treatment. Specifically, one study reported a negative relationship between overall QoL and the five domains of the SCNS (Giuliani et al., 2016) and one reported a relationship with physical QoL across four domains (Sanders et al., 2010). Similarly, the unmet needs across the five domains of the CaSUN were associated with poorer QoL (Molassiotis et al., 2017). For two of the remaining studies, the relationship between unmet need and QoL was limited to physical and/or psychological domains of the SCNS (Chan et al., 2012; Zhu et al., 2018). The latter two studies were conducted in China/Hong Kong, and it could be speculated that the difference reflects either a cultural difference or a difference in health care provision.

The findings tentatively suggest that unmet physical and psychological needs may have the most impact on QoL and reflect the high symptom burden (Li & Girgis, 2006) and psychological distress (Brown et al., 2014) associated with lung cancer. While the SCNS does not directly assess some of the potentially distressing symptoms frequently experienced by people with lung cancer (e.g., breathlessness), three studies reported an association between unmet needs in the physical domain and QoL (Chan et al., 2012; Giuliani et al., 2016; Sanders et al., 2010). Previous research with lung cancer patients has also indicated that declines in QoL are associated with increased symptom severity (Hung et al., 2018; Lowery et al., 2014) particularly when the symptoms, such as fatigue, interfere with daily functioning (Shi et al., 2016). Lack of energy and tiredness were among the most highly endorsed unmet needs in the physical domain in the current review. This is consistent with findings from other lung cancer research; specifically, fatigue is thought to be present in 57%–100% of patients with a considerable impact on QoL (Carnio et al., 2016). Taken together, these findings indicate that lung cancer has a considerable impact on survivors' ability to continue with everyday activities,with concomitant negative affect on their QoL.

Four studies reported an association between unmet psychological needs and poorer QoL (Chan et al., 2012; Giuliani et al., 2016; Sanders et al., 2010; Zhu et al., 2018). Uncertainties about the future, fears and worry were among the most common unmet needs within the psychological domain of the SCNS in the current study. In addition, fear about the cancer returning was one of the most endorsed items in the study using the CaSUN (Molassiotis et al., 2017). This is probably not surprising, as the poor prognosis and fears about recurrence frequently associated with lung cancer (Hsieh et al., 2017) may influence how the patient responds to, and interprets both diagnosis and subsequent treatments (Brown et al., 2014). In addition, changes to treatment and potential disease progression may contribute to feelings of uncertainty and lack of control (Borneman et al., 2014). Unmet support needs from healthcare professionals have been associated with poorer QoL in other studies with lung cancer patients (Luszczynska et al., 2013); importantly, such support could help to mitigate some of the distress associated with uncertainty and psychological symptoms such as anxiety and depression (Liao et al., 2011).

While there was a high level of unmet needs within the healthcare system and information domain of the SCNS, this was only associated with QoL in two of the studies included in this review. However, this domain is similar to the comprehensive cancer care and information domains of the CaSUN which were negatively associated with QoL (Molassiotis et al., 2017). Patients are likely to vary in terms of what and how much information they wish to have (Hansen et al., 2012), and information needs may change over time, for example when different treatment modalities are introduced (Okediji et al., 2017). Interestingly, patients who had their information needs met in one of the included studies reported better QoL than those who had not (Pardon et al., 2011). This information may have helped to reduce uncertainty about the future. However, as the authors pointed out, this does not suggest a causal relationship, as physicians may be more likely to give such information to patients who are perceived to have a better QoL in the first place (Pardon et al., 2011).

The fewest unmet needs were revealed in the sexuality domain of the SCNS and the relationships domain of the CaSUN. This is similar to other findings (Moghaddam et al., 2016) and may be explained by either a reluctance to express this need or a reduced salience during treatment and recovery. Nevertheless, needs are likely to be interconnected, and unmet needs in this domain may, for example, contribute to unmet needs in the psychological domain (Okediji et al., 2017). Similarly, meeting some informational needs may help to resolve psychological needs such as uncertainty and physical needs such as management of fatigue (Moghaddam et al., 2016). The likely interplay between unmet needs suggest that any intervention may need to take an interdisciplinary approach to improve well-being (Borneman et al., 2014).

4.1 Study limitations

While we conducted an extensive search, only six relatively small studies were included in the review. We limited our search to the last 12 years as older studies may not reflect changes in cancer care delivery and current practice in supportive care services. It was not possible to include potentially eligible data from studies with mixed cancer types (Hansen et al., 2012; Pisu et al., 2018) as we were unable to access the disaggregated data. In addition, the majority of the included studies were conducted in the Asia-Pacific region and any possible cultural differences in health care systems and the expectations related to family responsibilities in collectivist countries such as China (Lee et al., 2014) may limit the generalisability of the findings.

The included studies were cross-sectional in design, and therefore, we could not draw any conclusions about causal relationships or predictive direction. None of the studies included in this review used lung cancer specific measures of QoL, such as the FACT-L, and the EORTC lung cancer module (LC-13). Three studies were part of larger projects with mixed cancer types and thus needed a more generic tool (Chan et al., 2012; Molassiotis et al., 2017; Zhu et al., 2018). In addition, only two studies used the full versions of the QoL questionnaires. The use of single items or subscales may relate to concerns regarding perceived response burden and questionnaire length (Rolstad et al., 2011), although this was not explicitly explained in the included studies. The variety of QoL assessment tools and the use of single items in the included studies meant that we were unable to fully examine the multidimensional aspects of QoL and how these may be related to unmet needs. For example, a systematic review of social support among lung cancer patients and QoL was able to determine that perceived support from healthcare personnel was associated with several aspects of QoL (emotional, physical, functional and global index), while support from family and friends was associated with emotional aspects of QoL (Luszczynska et al., 2013). It is possible, in the context of the current review, that unmet needs in the physical domain of the SCNS would be related to the physical and functional aspects of QoL.

4.2 Future directions

This review indicated that higher unmet needs are associated with decreases in QoL, particularly those related to physical and psychological needs, in patients with lung cancer during the survivorship period. While some needs may be present along the continuum of treatment (Maguire et al., 2013), it seems likely that individual patient needs change over the course of the disease trajectory, from diagnosis through to survivorship or end-of-life (Fitch et al., 2008; Huang et al., 2014). Therefore, it has been recommended that needs assessments are conducted at key transitions, such as soon after diagnosis, the start and completion of treatment and the onset of any new symptoms or recurrence (Detterbeck et al., 2013).

Nearly two thirds of the patients in this review had been diagnosed with advanced cancer (Stage III or IV), which may account for the high level of needs reported. The respondents were at different stages of the lung cancer trajectory, from the newly diagnosed to those who had completed treatment. This makes it difficult to draw firm conclusions as to any possible dynamic changes in unmet needs and QoL. Longitudinal studies focusing on assessments of need at transitional stages could help to provide clinically meaningful information so that the provision of supportive care is adapted to any changing needs (Maguire et al., 2013). A reduced QoL may persist for some time after treatment (Huang et al., 2014), and as more people survive lung cancer, there is an imperative to also consider longer-term survivorship needs of this population (Al Achkar et al., 2020). There is also a growing need for both cross-sectional and intervention research that includes caregivers throughout the lung cancer survivorship trajectory (Lehto, 2017).

留言 (0)

沒有登入
gif