Unmet care needs of advanced cancer patients and their informal caregivers: a systematic review

Background This systematic review aimed to identify the unmet care needs and their associated variables in patients with advanced cancer and informal caregivers, alongside summarizing the tools used for needs assessment. Methods Ten electronic databases were searched systematically from inception of each database to December 2016 to determine eligible studies. Studies that considered the unmet care needs of either adult patients with advanced cancer or informal caregivers, regardless of the study design, were included. The Mixed Methods Appraisal Tool was utilized for quality appraisal of the included studies. Content analysis was used to identify unmet needs, and descriptive analysis was adopted to synthesize other outcomes. Results Fifty studies were included, and their methodological quality was generally robust. The prevalence of unmet needs varied across studies. Twelve unmet need domains were identified in patients with advanced cancer, and seven among informal caregivers. The three most commonly reported domains for patients were psychological, physical, and healthcare service and information. The most prominent unmet items of these domains were emotional support (10.1–84.4%), fatigue (18–76.3%), and “being informed about benefits and side-effects of treatment” (4–66.7%). The most commonly identified unmet needs for informal caregivers were information needs, including illness and treatment information (26–100%) and care-related information (21–100%). Unmet needs of patients with advanced cancer were associated with their physical symptoms, anxiety, and quality of life. The most commonly used instruments for needs assessment among patients with advanced cancer were the Supportive Care Needs Survey (N = 8) and Problems and Needs in Palliative Care questionnaire (N = 5). The majority of the included studies investigated unmet needs from the perspectives of either patients or caregivers with a cross-sectional study design using single time-point assessments. Moreover, significant heterogeneity, including differences in study contexts, assessment methods, instruments for measurement, need classifications, and reporting methods, were identified across studies. Conclusion Both advanced cancer patients and informal caregivers reported a wide range of context-bound unmet needs. Examining their unmet needs on the basis of viewing patients and their informal caregivers as a whole unit will be highly optimal. Unmet care needs should be comprehensively evaluated from the perspectives of all stakeholders and interpreted by using rigorously designed mixed methods research and longitudinal studies within a given context.


Background
According to the World Health Organization (WHO), more than 15 million people will be diagnosed with cancer by 2020 [1]. With the advances in cancer treatments, the illness trajectory and prognosis of cancer have changed, and patients diagnosed with advanced cancer can live for a relatively long period [2,3]. However, lengthy cancer experience and anticancer treatments make patients suffer from a wide range of problems, such as physical, psychological, emotional, and practical issues [4]. Cancer-related symptoms and patients' experiences during cancer treatment vary across different cancer stages, and patients at advanced stage commonly experience different symptoms from those with early-stage cancer [5,6]. Such 'chronic and uncertain' conditions pose a challenge to not only the cancer services but also to their informal caregivers [7]. Informal caregivers commonly take care of their loved ones for a long period [8]. The long-term caregiving process is physically and psychologically challenging, particularly when taking care of patients with advanced cancer [9]. Many informal caregivers, including those who do not regard caregiving as a burden, suffer from a wide range of problems, such as sleep disturbance, anxiety, depression, and practical and financial difficulties [10,11]. Informal caregivers are usually regarded as fellow sufferers alongside patients [12]. Unmet needs of patients can increase the level of caregiver burden [13]. In turn, caregivers' problems are closely linked with patients' well-being [14], and unsolved problems or unmet needs of caregivers will not only decrease their own quality of life [15] but also affect the patients' health outcomes negatively [15]. Informal caregivers and patients with advanced cancer are considered a whole unit in fighting the illness [10].
High-quality and patient-and-family-centered care is needed to address the problems of both the advanced cancer patients and their informal caregivers, including symptom and side effect management, as well as emotional, psychosocial, and spiritual support. All these aspects of support are typically categorized under palliative care [16]. Mismatched healthcare that is inconsistent with patients and caregivers' needs can increase healthcare expenditure and lead to harmful effects [17]. Therefore, the unmet care needs of patients and informal caregivers should be comprehensively assessed prior to designing and providing tailored palliative care services [18,19]. Care needs are defined as "the requirement of some action or resource in care that is necessary, desirable, or useful to attain optimal well-being" (Foot, 1996, as cited in Sanson-Fisher, et al., 2000, p.227) [20]. Unmet needs assessment is designed to identify how well and how much their needs have been satisfied or not [21]. An early review [17] summarized the instruments for needs assessment; however, a majority of these instruments have been designed for general patients with cancer (e.g., Supportive Care Needs Survey, SCNS [17]). After the publication of that review, several tools that were specifically designed for advanced cancer patients (e.g., Needs Assessment of Advanced Cancer Patients, NA-ACP [22]) have been developed and used.
An early systematic review [21] published in 2009 analyzed the unmet needs of patients with advanced cancer with nine included studies. Another systematic review [7] with 23 studies reported eight unmet need domains. These two systematic reviews only focused on patients, with limited literature searches in only four databases. Meanwhile, the inclusion criteria were relatively ambiguous in the second review because studies with mixed samples (patients at different cancer stages) were included; moreover, the definition of advanced cancer was not presented [7]. Moreover, neither of the two reviews summarized and reported detailed information regarding the needs assessment tools used, which is important information to allow readers to appreciate the quality and reliability of study results. Furthermore, to date, no systematic review has been conducted to explore the unmet needs of informal caregivers of patients with advanced cancer. Therefore, the current systematic review was carried out to update evidence from previous reviews and provide a more comprehensive picture regarding the unmet needs among patients with advanced cancer and informal caregivers. An intensive literature search was performed on 10 electronic databases, and the inclusion criteria were more specific for advanced cancer diagnosis than those of the previous reviews. This current systematic review also included informal caregivers on the basis of the following concepts: fellow sufferers [12], a whole unit [10], and patient-and-family-centered care that is emphasized by the WHO [16]. Specific objectives of this review included: (1) to identify the unmet care needs and their associated factors in patients with advanced cancer and their informal caregivers, and (2) to summarize needs assessment tools that were used in the included studies.

Search strategies
With consideration of the language expertise of the review authors, English and Chinese databases were included. Ten databases, including PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL), EMBase, Cochrane Central Register of Controlled Trials (CEN-TRAL), PsycINFO, Web of Science, Wan Fang Data, China National Knowledge Infrastructure (CNKI), Chongqing VIP (CQVIP), and Chinese Biomedical Literature Database (CBM), were searched systematically from inception of each database to December 2016. Restrictions regarding study design were not set. The used MeSH terms, key words, and free words included needs assessment, assessment of healthcare needs, unmet needs, neoplasms, advanced cancer, terminal cancer, metastatic cancer, and the forth. Manual searches were also conducted by examining the reference lists of the included studies. Three representative search strategies of this systematic review are listed in Table 1.

Study identification and data extraction
Duplications were identified and eliminated through a reference management software (NoteExpress). Titles and abstracts of the remaining studies were screened independently by two review authors (WT and TJY), and full text of potentially eligible studies were subsequently located for further screening. Studies satisfying the following inclusion criteria were included: (1) studies that included either adult (≥18 years old) patients with advanced cancer 1 or adult informal caregivers of patients with advanced cancer; (2) studies that reported data in terms of unmet care needs 2 or concerns that are directly linked to the unmet care needs of patients with advanced cancer and/or their informal caregivers, regardless of the study design; and (3) accessible full texts were published in peer-reviewed journals. Exclusion criteria were: (1) studies with mixed sample of patients with cancer at any cancer stage (except those patients with advanced cancer who were analyzed separately); (2) studies solely focusing on quality of life [21], satisfaction with healthcare services, care service utilization, or presence of symptoms/problems; (3) studies focusing on instrument development, translation, or evaluation; and (4) conference articles with only abstracts, editorial comments, guidelines, policies, or treatment recommendations. Data were extracted by two independent review authors. These data included information regarding the first author of the study, year of publication, country of origin, research setting, research design, sampling approach, sample size, need assessment methods (interview or other instruments), prevalence of unmet needs, and related factors for unmet needs. Any disagreement was settled and discussed by the two other review authors (CPM and AM).

Methodological quality appraisal
The methodological quality of included studies was assessed by two review authors (WT and TJY) independently with the Mixed Methods Appraisal Tool (MMAT) [25]. This tool is highly efficient; it takes approximately 14 min to evaluate one study [25] with robust consistency among reviewers (intraclass correlation = 0.72 [25]); MMAT is specifically designed to assess the quality of either quantitative or qualitative studies. Four different quality criteria for qualitative studies and different types of quantitative studies, including randomized control trials, quantitative nonrandomized trials, and quantitative descriptive studies, were used [25]. Each criterion was graded as 0 (unmet) or 1 (meet), and the global score of each study was calculated from 0 to 4 (0 = no criterion satisfied, 1 = satisfied one criterion, 2 = satisfied two criteria, 3 = satisfied three criteria, and 4 = satisfied all four criteria). When any disagreement occurred, the review authors conducted a group discussion to reach final agreement.

Data analysis
Content analysis [26] was used to identify the unmet need domains of patients with advanced cancer and informal caregivers across quantitative and qualitative studies. A priori content categories of patients with advanced cancer were determined on the basis of previous studies; these categories included health system and information, patient care and support, activities of daily living (ADL), physical, psychological, financial, and spiritual [7]. With regards to informal caregivers, five content categories were determined on the basis of a previous review [10]; these categories included cancer care services, informational, psychological, spiritual, and social needs. Data of the included studies were compared, combined, and clustered with respect to those domains for patients and informal caregivers. Terms, such as instrumental and personal care, were included in the ADL domain because they were frequently mentioned in several North American studies [21]. Summative content analysis was used to identify and extract new categories within content not covered by previous domains. The approach of descriptive analysis was used for the prevalence of unmet needs due to the significant heterogeneity of the included studies [27]. Variables associated with patients and informal caregivers' needs and used instruments were analyzed through descriptive approach.  [9,30,32,35,39,40,49,51,52,57,62,64] explored the unmet needs of informal caregivers, and five other [48,56,59,63,67] studies investigated the unmet needs of patients with advanced cancer and their informal caregivers. With regards to sample sources, six studies [32,40,45,46,49,61] reported no information regarding the recruitment setting, while in the remaining studies patients, and/or caregivers were mainly recruited from outpatient departments (n = 16), inpatient departments (n = 11), home/home-based care units (n = 10), and mixed settings (n = 7). In terms of cancer sites, 29 studies focused on patients with mixed cancer site and/or their caregivers, 11 studies focused on specific patients with cancer and/or caregivers (3 studies on prostate cancer [57,69,73], 5 studies on breast cancer [41,48,58,60,75], and three on lung cancer [35,42,71]), while 10 other studies [47,50,51,53,59,64,66,68] reported no information about cancer types. The diagnostic criteria of advanced cancer were presented in 13 studies (13/50), with five studies [6,30,31,60,61] adopting the criteria of cancer with metastasis, and seven studies [9,41,42,45,58,63,75] using the stage III/ IV criterion according to TNM staging system. With regards to geographic distribution, nine studies were conducted in the USA [38,40,46,49,52,57,59,70,74], seven were in mainland China (six of which were conducted in Shanghai) [9,53,[63][64][65][66][67], five in Australia [6,54,55,60,68], five in the Netherlands [29-31, 34, 44], four in Canada [47,50,56,73], three in Japan [33,39,41], three in Taiwan [35,42,62], two in the UK [69,71], two in Denmark [45,72], two in Hong Kong [58,75], and one each in Italy [28], France [61], South Korea [32], Spain [37], Indonesia [36], Czech Republic [43], India [51], and Bangladesh [48]. Characteristics and main findings of all included studies are presented in Table 2.

Quality of the included studies
The methodological quality of the included studies was generally robust, with 17 and 18 studies satisfying all four criteria (34%) and three of the four criteria (36%), respectively. The prominent weaknesses of 43 quantitative studies were poor sampling strategy and low response rate. The response rates of 16 studies [32, 33, 37, 39, 40, 43, 47, 52-56, 61, 63, 68, 74] were lower than 60%, and 14 studies [30-32, 38, 42, 43, 51-54, 67, 68, 70, 74] failed to report the sampling method, sampling procedure, or sample size justification. Among the seven other qualitative studies, three studies (3/7, 42.9%) [49,62,73] failed to interpret how findings were related to the study context, and two studies (2/7, 28.6%) [57,73] provided no explanation on how the research process was influenced by the researchers. The overall quality score of each study is presented in the first column of Table 2.

Descriptions of unmet needs in patients with advanced cancer
A total of 12 domains of unmet needs were identified from 34 quantitative and 4 qualitative studies. These domains included physical, ADL, psychological, health system and information, patient care and support, social, communication, financial, spiritual, autonomy, sexuality, and nutritional needs.

Unmet patient needs based on quantitative studies
Study sample sizes ranged from 40 to 977, with the average sample size being 165 and the response rate ranging from 36 to 100%. Physical needs were reported in 24 studies, and the most prominent physical unmet need was fatigue [6,31,33,34,42,43,45,47,50,54,56,63].
According to four qualitative studies [48,49,57,62], three similar unmet need domains, namely, informational, psychological, and social needs, were identified through summative content analysis. Informal caregivers commonly stated about "unmet information needs in terms of disease, treatment, side effects, care services, symptom management, nutrition, medication and nursing aids" (informational) [48,57,62], "feelings of sadness and loneliness, as well as a sense of abandonment, fear and helplessness" (p. 147) [48] or "insufficient listening and encouragement from other family members and professionals" [62] (psychological), and "feelings of isolation due to the lack of social activities" (social) [57]. Several specific unmet needs, including the manner of communication between professional staff and caregivers or patients, the administration and function of the healthcare system, and some practical assistance, such as cleaning the house and walking the dog [49], were also identified in qualitative studies [49].

Variables associated with the unmet needs of patients with advanced cancer
Variables associated with the unmet needs of patients with advanced cancer are summarized in Table 4. Relevant variables were categorized as patient-related variables (demographics, disease-related, physical, and psychological) and informal caregiver-related variables (age, gender, and psychological distress of informal caregivers).
In several studies, age, gender, marital status, education level, and income level were insignificantly associated with patients' unmet needs. Although a significant relationship was reported, results were inconsistent across studies in terms of age and marital status. With regards to gender, three studies [28,33,63] reported that female patients indicated more physical and psychological unmet needs than those of male patients. Patients who were living alone experienced high psychological needs [28], and patients with high educational level presented considerable unmet needs in physical [42], ADL [42], information [44], community service [46], and sexuality [58] domains. Moreover, financial needs were less reported in patients with high income [46,63]. Notes: Needs items (sentences or phrases) which were put in the quotation marks were directly extracted from the corresponding included studies  "fin": financial needs; "PM": pain management; "soc": social needs; "phy": physical needs; "psy":psychological needs; "psys": psychosocial needs; "inf": information needs; "com": communication needs; "occup": occupational needs; "HSIPS": health system, information, and patient care support; "med": medical needs; "spiri": spiritual needs Four studies [42,58,70,75] explored the relationships between symptom distress and unmet needs, and all these studies showed that patients with symptom distress experienced more unmet needs in the psychological, physical, and ADL domains. Patients with poor ability in daily living [28] indicated more unmet needs than those of independent patients, especially in terms of information, communication, psychological, and occupational needs.
Two studies [28,70] showed that no relationships were observed between cancer site and their unmet needs, but two other [46,63] studies showed opposite results. Two [42,75] out of five studies reported that no relationship was observed between cancer stage (only stages III and IV) and unmet needs, and three ones [58,63,70] indicated that patients with stage IV cancer presented more unmet needs than those with stage III cancer. Results were inconsistent across studies for cancer treatment, with two studies showing no relationship [42,44] and two other studies suggesting either positive [75] or negative [58] relationship.
Patients with anxiety experienced high levels of physical, psychological, healthcare, and information, as well as ADL unmet needs, which was confirmed across several studies [33,41,42,44,58]. Patients with depression [42,44,58,70] demonstrated varied results. Patients with low quality of life showed high unmet needs, especially in physical and psychological domains [41,43]. Patients reported more unmet needs when their caregivers were male [28], young people [28], or those who suffered from psychological distress [28].

Variables associated with the unmet needs of informal caregivers
Older caregivers [30,35] showed less unmet needs in terms of financial, social, and care-related information needs than those of younger caregivers. Caregivers in different caregiving settings reported different levels of unmet needs (home>general hospital>hospice care unit) [32,39]. Caregivers with many physical problems experienced many unmet needs [35,63]. Caregivers had higher levels of unmet needs when patients suffered from anxiety [35], depression [35], or low physical performance [35]. Results varied across studies in terms of gender [30], length of caregiving [9,63], and education level of caregivers [63] (Table 5). Similarly, results were conflicting with regard to the relationships between caregivers and patients. One study [39] showed that spousal caregivers presented many information needs, and another study [63] indicated that non-spousal caregivers reported many unmet needs.

Discussion
The included studies highlighted that both advanced cancer patients and their informal caregivers possess a wide range of unmet needs. Psychological and physical unmet needs are two areas of focus for patients with advanced cancer; this result is consistent with a previously published review [7]. Among informal caregivers who had experience in managing patients' negative emotions, more than 30% of them reported that emotional management is the most challenging part of caregiving [76]. Three other unmet needs, namely, the need for autonomy, communication, and nutrition, were identified in this review compared with the previous review [7]. These needs may be related to the differences in cultural contexts, healthcare systems, and economic levels because several included studies in this review were conducted in eastern and developing countries. For instance, the need for autonomy is commonly culture-related [36]. Family members usually make decisions for patients in eastern cultures because family-collective decision-making is much more popular there than in other cultures [77]. This result showed the importance of developing tailored healthcare services or interventions based on context-specific unmet needs. Disease-related information needs were the most commonly reported unmet needs of informal caregivers. Considerably fewer studies reported unmet needs that are associated with the caregivers' own well-being, as they generally focus more on the patients' well-being than their own [30]. The prominent care needs of each domain were identified for patients with advanced cancer and informal caregivers in this review provide useful information and evidence for the development and implementation of tailored healthcare services. For example, emotional support was identified as the most commonly unmet need in the psychological domain for patients, thereby indicating that emotional distress (e.g., anxiety and depression) management should be a priority when providing mental health services. In addition, patients with advanced cancer and informal caregivers' unmet need domains involved multiple disciplines, which indicated that healthcare services should be multidisciplinary. The value of multidisciplinary care for patients with cancer has been well recognized [78]. Support for informal caregivers is suboptimal in many instances [79]. The unmet needs of informal caregivers are often ignored and excluded from healthcare planning [80,81].
The prevalence of unmet needs varied across the quantitative studies for both patients and caregivers. This variability may be caused by the heterogeneity of the included studies, which were conducted within different cultural contexts, healthcare systems, and economic levels that may be associated with unmet needs. High-income countries or regions generally present well-established healthcare service systems, which can facilitate the timely identification and resolution of healthcare problems (several physical symptoms particularly require high-quality professional support [28]). Different study designs, especially the diverse instruments used, for unmet needs assessment also contribute to this heterogeneity. The highlighted heterogeneity makes it difficult to gauge and pool the percentages of unmet needs by domains. SCNS was the most commonly used instrument, which was used in eight studies. However, these eight studies adopted five different variants of the same scale, with 13 [61], 33 [58], 34 [6,33,41], 59 [60], and 61 items [50] for each of the five versions. Different methods of need classification are also a major barrier in gauging unmet needs by domains. For instance, in SCNS, several items were classified as spiritual needs (e.g., [50]). In other studies, the same items were coded as psychological needs (e.g., [41]). Moreover, approaches in defining unmet needs were inconsistent. Among studies that utilized the SCNS, several of them regarded moderate and high levels of need as unmet needs (e.g., [41]. In other studies, low need level was calculated as an unmet need (e.g., [50]). Different reporting methods also caused heterogeneity. Several studies reported the prevalence of unmet needs by domains without specifying the percentage of items within each domain. Some studies (e.g., [33]) only listed the prevalence of the top 10 or 20 items without reporting the prevalence by domain. Thus, directly combining the prevalence of reported items within a domain may increase the risk of overestimating the actual unmet need level [21].
Although consistent results across studies showed that patients with advanced cancer with symptoms of distress and anxiety and low quality of life are more likely to report high demands of unmet needs, the conclusion must be interpreted with caution. Causality cannot be established because almost all of the included studies were cross-section in design. Other patient-related variables with inconsistent results, (e.g., gender, marital status, education level, cancer site, and depression) may be caused by cultural differences and/or methodological flaws (e.g., insufficient sample size to explore relationships between two factors) of the included studies. Hence, more longitudinal studies with rigorous study designs should be adopted. In addition, whether caregivers' health outcomes were associated with the unmet needs of patients is still unclear because of the limited evidence that can be drawn from current studies. Therefore, more studies should focus on caregiver-related variables. Relevant studies regarding variables associated with informal caregivers' unmet needs are limited, and no conclusion can be drawn from the current findings.
Patients with cancer at an advanced stage commonly experience fluctuating unmet needs over time due to rapid disease progression [6]. Nevertheless, little is known about how patients with advanced cancer and/or their informal caregivers' unmet needs change across the illness trajectory. Almost all the included quantitative studies investigated unmet needs at a single time point with cross-sectional study designs. Unmet care needs assessment in the majority of the included studies is also mainly problem-oriented from a biomedical lens. Few studies considered contextual issues (sociocultural and healthcare service provisions) when assessing and interpreting results in a given context although it will be of benefit to the development and implementation of tailored interventions at a local level. Accordingly, qualitative studies are an appropriate approach because it can explore participants' in-depth experience and subjective feelings that cannot be measured by quantitative methods; additionally, the scope can be much broader than those of quantitative methods [82,83]. Deeper understanding of unmet needs can be extracted from the qualitative studies than from quantitative findings. However, limited studies adopted qualitative study designs, and only few studies utilized mixed methods. Care needs should be comprehensively evaluated from all stakeholders, including patients, caregivers, and healthcare providers [84]. A comprehensive understanding of both patients with advanced cancer and informal caregivers' unmet needs can enable healthcare providers to develop evidence-based and tailored interventions [18]. Nevertheless, the majority of the included studies assessed patients' unmet needs only, and almost all included studies examined unmet needs from the participants' own perspective rather than from the perspectives of all relevant stakeholders. Despite that the concept of patient-and-family-centered care is advocated by the WHO [16], structured unmet needs assessment of informal caregivers is still an uncommon practice. Only a few studies assessed the unmet needs of patients and informal caregivers, and their unmet needs were assessed separately. The mechanism of integrating the data of patients and caregivers should be considered to further embody the conceptualization as a whole unit. Focused group with mixed samples, including patients and informal caregivers in the same group, may be an appropriate approach. Finally, research instruments used for needs assessment in several included studies were inappropriate. Some scales are generic ones used for supportive care needs assessment. Several items, such as "fear about the cancer spreading," may be unsuitable for patients with advanced cancer.
A strength of this systematic review is that a large number of studies with considerable information were assimilated and analyzed through a systematic method, which can minimise biases and facilitate reliable conclusions. This work is the first systematic review conducted by considering patients with advanced cancer and their informal caregivers as a whole unit. However, this review also presents several limitations. First, subgroup analysis in terms of contexts and economic levels was not conducted. Second, given the confounding factors and insufficient number of studies in each subgroup, meta-analysis was also not performed to compare the prevalence of each identified need domain. Third, language bias cannot be excluded because only papers that were published in English or Chinese language were included. Finally, instruments for needs assessments were only summarized from the included studies, and studies in terms of instrument development were excluded.

Conclusions
A wide range of unmet care needs existed in both advanced cancer patients and informal caregivers. Given the context-bound feature, their unmet needs should be comprehensively assessed and interpreted from the perspectives of all stakeholders within a given context by using rigorous mixed methods research and longitudinal research with prospective study designs. Assessing unmet care needs by viewing patients with advanced cancer and their informal caregivers as a whole unit is highly desirable. Associated factors of their unmet needs should not be ignored, which can provide evidence for decision-making with regards to healthcare resource allocation. The value of better examining unmet needs