We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (Page et al., 2021). The protocol of this systematic review has been registered in the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42020188560). The University Medical Center Groningen (UMCG) team (TI, AL, MH, and SD) established the search strategy. A senior researcher (JW) from Uppsala University then reviewed it following the PRESS peer review guidelines (McGowan et al., 2016). SD searched the electronic databases PubMed, CINAHL, and PsycINFO) in October 2019 for publications in the period of January 2010 to October 2019. Later the search was extended till November 2021. We created a search string using the free text words and MESH terms when available. We combined the three main categories with the Boolean operator “AND” to identify the relevant articles. The categories were: the informal caregiving population (e.g., caregiving, caring), needs (e.g., needs, needs assessment), and family relationships (e.g., spouses, children). We used the following search string in the PubMed database:

((((“Caregivers”[Mesh] OR caregiv*[tiab] OR ((“Family”[Mesh] OR family[tiab] OR spous*[tiab] OR parent*[tiab] OR husband*[tiab] OR wife[tiab] OR wives[tiab] OR partner*[tiab] OR adult child*[tiab]) AND caring[tiab])))) AND ((“Health Services Needs and Demand”[Mesh] OR “Needs Assessment”[Mesh] OR needs[tiab]))) AND ((“Spouses”[Mesh] OR “Siblings”[Mesh] OR wife[tiab] OR wives[tiab] OR husband*[tiab] OR marital[tiab] OR spous*[tiab] OR sibling*[tiab] OR adult child*[tiab] OR brother*[tiab] OR sister*[tiab] OR daughter*[tiab] OR son[tiab] OR sons[tiab] OR granddaughter*[tiab] OR grandson*[tiab] OR grandchild*[tiab])).

In addition to the advanced search on the electronic databases, SD performed a reference check of the final shortlisted articles to identify the relevant studies that were not found in the database search.

Two authors, SD and GF, independently performed the selection of studies in two phases. In the first phase, the titles and abstracts of all articles found in the electronic database search were screened. The author (GF) screened 10% of the retrieved articles in the first phase (Gough et al., 2017). To be included, studies needed to meet the following criteria: (i) they provided data on the needs or unmet needs of one or multiple groups of informal caregivers, i.e., ICGs taking care of a spouse, parent, adult child, adult sibling, grandparent or grandchild; (ii) had a qualitative, quantitative or mixed-method study design, (iii) were published in English; and (iv) included adult ICGs, i.e., aged 18 years or above. Studies were excluded if (i) they presented outcomes in which data of one group of ICGs were mixed with the data of another group of ICGs or (ii) were published as gray literature (i.e., conference abstracts, presentations, proceedings; unpublished trial data; government publications; and reports such as white papers, working papers, and internal documentation).

In the second phase, we screened the shortlisted articles for full text. The author (GF) again screened 10% of the shortlisted articles in the second phase. The definition of need was kept broad, including personal (e.g., need for leisure time) and care-related needs (e.g., need for caregiving support) of the ICGs. Articles describing an evaluation of an intervention program (i.e., describing what ICGs need in that specific program) were not considered. In this study, we define spousal ICGs as caregivers who were in an intimate relationship with their CR; they could be either married, living in a partnership, or unmarried, irrespective of whether they share the same household or not. The studies were excluded when both authors (SD and GF) were convinced that the study did not meet the inclusion criteria. Authors SD and GF experienced disagreement in 5% of the total number of articles screened by GF in the first phase and 2% in the second phase. These discrepancies were resolved with a detailed discussion between the researchers. If no consensus was reached, the articles were screened and discussed by the other authors to reach a consensus.

SD and GF independently extracted the data from the selected studies. The researcher (GF) extracted the data from one-third of the selected studies. The following data were extracted and recorded in a table: author name, year of publishing, country, study aim, study design, sample characteristics, the relationship of the ICGs with their CR, illness of the CR, and type of needs expressed by the ICGs (Supplementary Table 1). We developed the table based on the Cochrane data collection form for intervention reviews on randomized controlled trials (RCTs) and non-RCTs (Li et al., 2021).

The quality assessment was performed by two authors, SD and GF independently for all the 22 selected studies. After completing the phase both the authors discussed the discrepancies in selecting the studies. The discrepancies were resolved with a detailed discussion between the researchers. If no consensus was reached, the articles were screened and discussed by the other authors (MH and SD) to reach a consensus. The quality of included qualitative studies was accessed using the Critical Appraisal Skills Program checklist (CASP) (CASP, 2017). The checklist consists of 10 questions, which assess qualitative studies on the following criteria: clarity of the aim, appropriateness of the study design, and the validity and implications of the findings. The first two questions were the screening questions, and if the answer to both was “yes,” the author proceeded with the remaining questions. All the questions were evaluated using the parameters “yes,” “no,” or “cannot tell.” “Yes” corresponded to strong quality, “no” to moderate, and “cannot tell” to weak quality with respect to the specific criteria (Table 1).

The quality of included quantitative studies was assessed using the Quality Assessment Tool for Quantitative Studies, recommended by Cochrane (Higgins et al., 2011). The quantitative studies were rated on selection bias, study design, confounders, blinding, data collection method, and withdrawals and dropouts (Table 2), using scores from one to three, where one indicated strong, and three indicated weak quality. Based on these criteria, the overall quality of the quantitative articles was rated as “1” (strong), if there were no weak ratings; as “2” (moderate), when one component had a weak rating; and as “3” (weak) when two or more components had weak ratings.

We used thematic analysis, a technique of narrative synthesis (Braun and Clarke, 2006), to synthesize the data for both qualitative and quantitative studies. The thematic analysis technique is commonly used for analyzing qualitative data but could also be used for quantitative data (Popay et al., 2006). Since the majority of articles provided qualitative data, thematic analysis was an appropriate method to describe and compare the main findings. We followed the convergent integrated mixed-method approach for synthesis and integration of qualitative and quantitative data (Stern et al., 2020). We carried out three steps for the thematic analysis. In the first step we performed data transformation for quantitative studies where we extracted data from quantitative studies and converted it into textual descriptions to allow integration with qualitative data. We then extracted line by line textual data for each included qualitative study article as well. The textual descriptions (qualitized data) from quantitative studies were assembled and pooled with the qualitative data extracted directly from qualitative studies. In the second step, the authors (SD, AL, GF, and MH) familiarized themselves with the extracted data on the needs of ICGs. They generated an initial list of codes for different types of needs from the data. In the third step, when all the data was coded, initial codes were combined into themes. We used a mind map to group the codes with different types of needs into main themes of needs. The themes were compared to identify overlapping and unique needs of spousal, adult child, and adult sibling ICGs.

Figure 1 presents the PRISMA flowchart, reflecting the number of included studies at each step. In total, 5,414 articles were retrieved from the electronic databases, of which 468 articles were reviewed in full-text. After reviewing the full-text articles, 22 studies were included for the analysis. No new articles were included based on the reference check of the included articles.

Supplementary Table 1 presents the characteristics and main findings of the included qualitative (n = 18) and quantitative (n = 4) studies. The included articles reported the needs of spousal ICGs, adult child and adult sibling ICGs aged 18 years or above. We found a limited number of articles that represented the relationships of parents, grandparents, and grandchildren; therefore, we did not include other relationships. Out of these 22 studies, (i) 11 articles reported the needs of spousal ICGs, (ii) five articles reported the needs of adult sibling ICGs (iii) three articles reported the needs of adult child ICGs, and (iv) three articles discussed the needs of both spousal and adult child ICGs with data analyzed separately. The sample sizes ranged from 6 to 139 ICGs for the qualitative studies and 68 to 862 ICGs for the quantitative studies. The studies were conducted in Australia, Canada, China, Denmark, France, India, Japan, Norway, Portugal, the Netherlands, United Kingdom, and United States.

The results of the quality assessment of qualitative studies are presented in Table 1. The assessment indicates that all the studies included clear aims, qualitative methods, recruitment strategies, data collection methods, and statements of findings (18/18). According to the information reported in the articles, most of the studies had considered ethical issues (11/18) and conducted rigorous data analysis (16/18). Concerns were noted in the potential for bias in the relationship between the researcher and participants as these articles did not report on the researcher’s role and it’s influence during data collection, with only 4/18 studies meeting the CASP criteria. The results of the quality assessment of quantitative studies are presented in Table 2. The assessment indicates that only the study design component was reported strong for all the articles, whereas other components were weak for one or more studies. Based on the assessment of all the components, two of the four quantitative studies were assessed as having moderate quality (Turner et al., 2013; Kobayakawa et al., 2016), and two have a weak quality (Peeters et al., 2010; Veil et al., 2013).

The assessment was conducted to improve transparency in the systematic review process. No study was excluded based on the evaluation.

The synthesis of the 22 included qualitative and quantitative articles resulted in seven themes of needs (Table 3). Out of the seven themes of needs, (i) four themes were reflected by all the three groups, i.e., a need for information, for support, for personal time, and to manage personal concerns, (ii) two themes were expressed only by the adult child and adult sibling ICGs, i.e., a need to be acknowledged and to be more than a carer, and (iii) one theme was expressed only by spousal and adult sibling ICGs, i.e., a need to maintain the relationship with the CR. The seven themes included certain sub-themes, which differed further for the spousal, adult child, and adult sibling ICGs. We will discuss the common and unique themes and sub-themes in detail below for the qualitative studies and quantitative studies.

All three groups indicated the need to know about the availability of services. This need was expressed by spousal and adult sibling ICGs in a qualitative study, and by adult child ICGs in a quantitative study, therefore the service needs for spousal and adult sibling ICGs were more descriptive in nature. They indicated the type of services they needed, such as home assistance and respite care, whereas the adult child ICGs only indicated a need to know about the services available and where to find them. The literature indeed suggests that ICGs lack awareness about the availability of services (Wiles, 2003; Bieber et al., 2019), and this lack of awareness acts as an important barrier that withholds ICGs to access and use these formal services and support (Innes et al., 2011; Bieber et al., 2019). Other barriers that withhold ICGs from accessing the services are lack of flexibility of services, lack of availability, or lack of support from the CR especially for respite care services (Fine and Thomson, 1997). It is important for ICGs to know about the availability of services as the literature suggests that ICGs who are unable to use the formal services and support often experience high levels of caregiver burden and poor health outcomes (Fine and Thomson, 1997).

All three groups expressed a need for social support from people around them and supportive care from the professionals. Interestingly, spousal and adult sibling ICGs expressed an additional need for financial support, which was not indicated by adult child ICGs. This need is in line with the literature that spousal ICGs experience higher financial strains than adult child ICGs and other relationships (Van Houtven et al., 2010; Pinquart and Sörensen, 2011; Lee and Zurlo, 2014). This could be because of the financial interdependence in a marital relationship, where the financial responsibility has to be taken care of by ICGs due to the illness of their spouse. Interestingly, adult sibling ICGs also expressed a need for financial support. There is limited literature to support the findings, but it could be linked to their transition from secondary to the primary caregiver in the later stage of life, where they express a need to be included in future planning for financial matters (Taggart et al., 2012; Davys et al., 2016).

The need to be acknowledged was expressed only by the adult child and adult sibling ICGs. Our findings are in line with the literature that suggests that spousal ICGs by default are acknowledged as caregivers by society because of the nature of their relationship with the CR (Lee and Smith, 2012). Adult child ICGs have the experience of caregiving, which is not recognizable by their peer group or family members, making it difficult for them to be understood or appreciated by the people around them (Nicholls et al., 2017). Adult siblings are not considered traditional caregivers, and their struggle of not being involved in caregiving by family members and health care workers is well discussed in the literature (Burke et al., 2015). They want to be involved by family members in future planning, especially in plans related to finance and legal matters (Heller and Arnold, 2010).

Adult child ICGs expressed an additional need to detach themselves from the responsibilities of caregiving. The two articles that indicated this need included a population of mostly young adult caregivers (YACs), the age of these ICGs ranged from 18 to 30 years. As we know from previous literature, young adults in this age group are in their transition period between being young and at the same time entering adulthood, establishing themselves with respect to certain aspects of life such as education, career, relationships, and social life. This period may be more challenging for YACs who have to integrate caregiving with other aspects of life (Haugland et al., 2020). We know from literature that caregiving has an impact on YAC’s academic performance, they get fewer opportunities to connect with their peers in college, maintain their relationships with friends and close ones, or start new relationships (Mickens et al., 2017). Therefore, it is imaginable that YACs experience a need to allow themselves to be young adults without thinking about caregiving responsibilities (Stephens et al., 2001; Broese van Groenou et al., 2013; Bastawrous et al., 2015).