Family reported outcomes, an unmet need in the management of a patient's disease: appraisal of the literature

Background A person’s chronic health condition or disability can have a huge impact on the quality of life (QoL) of the whole family, but this important impact is often ignored. This literature review aims to understand the impact of patients' disease on family members across all medical specialities, and appraise existing generic and disease-specific family quality of life (QoL) measures. Methods The databases Medline, EMBASE, CINHAL, ASSIA, PsycINFO and Scopus were searched for original articles in English measuring the impact of health conditions on patients' family members/partner using a valid instrument. Results Of 114 articles screened, 86 met the inclusion criteria. They explored the impact of a relative's disease on 14,661 family members, mostly 'parents' or 'mothers', using 50 different instruments across 18 specialities including neurology, oncology and dermatology, in 33 countries including the USA, China and Australia. These studies revealed a huge impact of patients' illness on family members. An appraisal of family QoL instruments identified 48 instruments, 42 disease/speciality specific and six generic measures. Five of the six generics are aimed at carers of children, people with disability or restricted to chronic disease. The only generic instrument that measures the impact of any condition on family members across all specialities is the Family Reported Outcome Measure (FROM-16). Although most instruments demonstrated good reliability and validity, only 11 reported responsiveness and only one reported the minimal clinically important difference. Conclusions Family members' QoL is greatly impacted by a relative's condition. To support family members, there is a need for a generic tool that offers flexibility and brevity for use in clinical settings across all areas of medicine. FROM-16 could be the tool of choice, provided its robustness is demonstrated with further validation of its psychometric properties. Supplementary Information The online version contains supplementary material available at 10.1186/s12955-021-01819-4.


Background
A person's chronic health condition or disability can have a huge impact on the quality of life (QoL) of the whole family. Sometimes this impact may be similar to or even greater than that experienced by the patient [1][2][3]. Although awareness of the impact of a person's disease on family quality of life (FQoL) has recently been increasing, there is a need to measure this impact in the clinical setting to inform those providing support to the family. Turnbull et al. first proposed the term in 2000 and defined normal "family quality of life" as being "where the family's needs are met, and family members enjoy their Shah et al. Health Qual Life Outcomes (2021) 19:194 life together as a family and have the chance to do things which are important to them" [4].
Golics et al. 's [5] detailed literature review of the impact of chronic disease on a patient's family revealed that various aspects of family life are affected by relative's health condition. That review only identified information about a few disease areas and specialities [5] and concluded that there was no generic instrument at that time to measure disease impact on family members of patients.
The investigation of FQoL is a newly emerging field, with research now extending to many different areas of medicine. It is, therefore, timely to update the existing knowledge base on the family impact of disease and identify the development of new generic and disease-specific FQoL tools. This critical appraisal of the literature builds on the areas covered by Golics et al. [5] and summarises the greatly increased research activity over the last seven years. It aims to identify the impact of chronic disease on family members of patients across a range of medical specialities and appraise the characteristics and measurement properties of existing generic and disease-specific FQoL measures.
The definition of 'family' has changed over time and its use is no longer restricted to describing 'two parents and their children living under the same roof ' . In this review, we use the term as defined by Poston et al. [6] as "People who think of themselves as part of the family, whether related by blood or marriage or not and who support and care for each other on a regular basis". This review studies the impact of a patient's disease on all family members, including partners, whether or not they are also carers. Although the terms family caregivers, carers and informal caregivers are often used interchangeably, the only caregivers covered by this review are those unpaid carers (caregivers) who are family members or partners.

Search strategy
A search strategy was developed to identify studies published up to January 2020 that reported the impact of chronic disease on patients' family members and partners. Six electronic databases were searched: Medline via OVIDSP; EMBASE via OVIDSP; CINHAL via EBSCO; ASSIA via ProQuest; PsycINFO Via OVIDSP; and Scopus using the PICO framework (Population: family members of chronic patients, Intervention: Patients chronic illness, Comparison: Non-applicable, Outcome: impact on family members) to identify and record the data (Additional file 1: Table S1a and S1b). The PICO framework was developed by the lead author and agreed by the other authors. The reference lists of included articles were also examined to ensure that all relevant articles were captured.
The search to identify existing generic and disease-specific FQoL measures was extended by combining search terms such as 'family*or caregiver' and 'quality of life' with the terms scale, index, measure, instrument, assessment, surveys, questionnaires, inventory, tools, generic or disease-specific (Additional file 1: Table S2). In addition, hand searches were carried out of the COnsensusbased Standards for the selection of health Measurement Instruments (COSMIN) [7] database and the reference lists of relevant articles. Google Scholar was searched for articles reporting development or psychometric properties of the instruments identified.

Eligibility criteria
Articles were included in the review if the source was an original paper, in the English language and measuring the impact of chronic illness or disability on patients' family members/partner using a valid tool. Studies were excluded if they were book chapters, congress abstracts, if they used qualitative methodology or if the caregiver was not a family member. This review paper is in two parts, the first part focuses on the impact of a patient's disease on family members and the second part appraises the instruments available to measure this impact. As one of the inclusion criteria for the second part was only to include quantitative techniques, it was felt methodologically appropriate to align the two parts by including only quantitative studies in the first part. We recognize this could be considered as a limitation of the study.

Screening
In the first stage of article screening, duplicates were removed, and irrelevant titles and abstracts were discarded based on eligibility criteria. In the second stage, full-text articles of potentially relevant abstracts were read and assessed against eligibility criteria by RS to make a final decision about study selection agreed by MSS and AYF.

Data extraction
Data extraction was carried out by RS and was discussed using an iterative process with other members of the research team (MSS and AYF). The data extracted included authors, publication year, country of study, study design, sample size, patients' chronic disease, family member gender, relationship to the patient, impact on the family members and tools used to measure this impact (Additional file 1: Table S3).
A separate data extraction table was used for recording psychometric properties of identified family QoL instruments.

Synthesis of data
We used a thematic approach to synthesise findings. Selected papers were carefully read by RS: in case of ambiguity, papers were discussed with FMA, AYF and MSS to ensure accuracy of data extraction. The data on the impact of patients' disease on family members were summarised as short notes for the 86 studies. These notes were then coded to capture their essence and finally, codes were sorted into potential themes.

Quality assessment and risk of bias
The quality of selected papers and assessment of risk of bias was evaluated using the Joanna Briggs quality assessment tool for cross-sectional and cohort studies, with the involvement of MSS and AYF [8]. The checklist consists of 8-11 questions with answers "yes", "no" and "unclear". When all answers were "yes", the study was considered to have less chance of bias and if any answer was "no"' the study was classified as having a risk of bias. The PRISMA principles were followed to ensure robustness of the review as well as minimising bias [9].

Screening
A total of 7,767 articles were identified. After removing duplicates and irrelevant titles, 558 abstracts were screened. The resultant 114 articles underwent full-text review, 86 articles met all inclusion criteria and were included in the final analysis ( Fig. 1).

Study characteristics
Eighty-one studies were cross-sectional, and five studies were longitudinal prospective cohort studies with follow-up ranging from one month to two years. The studies explored the impact of a relative's disease on a total of 14,661 family members, mostly 'parents' or 'mothers' , using 50 different tools across 18 specialities including neurology, oncology and dermatology and covering 33 countries including the USA, China, and Australia (Figs. 2 and 3; Additional file 1: Table S4 and S5). The most widely used tool to measure the impact of a patient's disease on a family member was the Zarit Caregiver Burden Scale (13 studies) followed by WHOQOL (11), SF-36 (11), SF-12 (nine), IOF (seven) and EQ-5D (six) (Fig. 2). While most of the articles reported the impact of a single chronic disease on family members, ten studies included more than one chronic condition, allowing comparison of the family impact of different diseases.

Quality assessment and risk of bias
Thirteen cross-sectional studies and one cohort study did not mention confounders and strategies to address them while one cohort study did not mention reasons for loss of follow-ups. However, the remaining requirements were met for all of these studies, which all fulfilled the minimum criteria for quality. None of the 86 studies was rejected based on their quality or risk of bias. Overall, all studies were moderate to high quality (Additional file 1: Table S6).

Synthesis of findings-key impact areas
This review revealed a huge impact of patients' illness on family members' QoL [10][11][12][13][14][15][16][17][18]. In general, relatives' chronic diseases impacted family members in similar ways, with some conditions such as cancer having a bigger impact than others. Some common themes identified in this review are discussed below.

Emotional and psychological impact
Caring for a relative's chronic disease affects family members' lives in many ways, impacting their emotional and psychological wellbeing [19]. The family members caring for their relative with a chronic disease were at risk of themselves developing a mental health condition, with an adult offspring or spouse at higher risk than other family members [10,20,21], and suffered similar psychological distress, depression and anxiety levels to that of the patient [22][23][24]. The presence of anxiety and/or depression in the family member was the most consistent factor influencing family members' burden and perceived health-related QoL (HRQoL) [25,26].

Nature of relationship and psychological impact
Mothers of children with chronic disease experienced high rates of stress, anxiety and depression [15,[27][28][29][30]. Parenting stress was higher when a child was of preschool age [31,32] and displaying disruptive behaviours and developmental disabilities [33] or showing flares due to increased severity of their condition [34,35]. Some parents perceived the increased caring demands of a sick child as 'intrusive' which led to higher levels of parental stress and psychological distress [36] affecting the perception of burden experienced by the mother [37]. However, this emotional distress did not result in mothers being less caring of the sick child [38]. The children of mothers with a chronic condition experienced more symptoms of hyperactivity and inattention, especially when the mothers had psychological problems [39]. Siblings of children with a more severe chronic condition and an unpredictable prognosis reported more internalising of problems and behavioural difficulties than siblings of children with a chronic condition that followed a daily routine treatment pattern [40]. However, poor emotional health of siblings of children with controlled asthma was not related to disease severity [41]. Moreover, what is worrying is that parents are sometimes unaware of the impact of their child's disease on their other children [42].

Gender differences
Female family members, spouses and mothers, experienced significantly higher rates of depression and anxiety than male family members [15,21,25,28,29,43] and the impact was greater when patients suffered from a severe disease such as a long-term mental health condition [44]. However, two studies showed fathers experiencing more stress [45] and lower HRQoL [46] than mothers. Such paternal outcomes could be explained based on increased stressors arising from disease flares, such as additional medical visits and medical bills, both of which could be particularly distressing for fathers compared to mothers [46]. The reverse gender difference was found in siblings of a patient, with female siblings experiencing a lower QoL than male siblings [40].

Impact on physical health
Caring for a relative with a chronic disease can have an impact on family members' physical health owing to the burden resulting from the relative's functional disabilities, cognitive impairment [27,47,48], medication management [49] duration of care [43,50] and total daily hours spent on assisting patients with basic activities of daily living and medical tasks [12,[50][51][52]. Caring for their relative can leave family members overwhelmed and physically exhausted [53,54], which may result in compassion fatigue. It is not the total number of years of caregiving that contributed to differences in compassion fatigue, but the number of hours per week [55], suggesting that intensity of caring rather than duration is the critical factor. Furthermore, family members of people with less severe chronic diseases reported only a moderate burden on QoL [56,57], indicating that caregiving burden is related to the severity of the patient's disease and the family member's perception of burden [35,58].
Meltzer et al. [61] found that parents of ventilatorassisted children experienced shorter sleep duration and greater variability in sleep quality impacting their physical health compared to parents of healthy children. In the mothers of children with Duchenne muscular dystrophy, impaired sleep quality was related to the disease duration [62], while the sleep disturbance in the parents of children with atopic disease was related to the children's sleep disruption [63]. The partners of cancer patients experienced poor sleep quality: there was a significant correlation between patients' and their partners' sleep quality and sleep onset latency [60]. Although partners used medication to minimise the negative impact of sleep problems, Chen et al. [60] argue that this could have affected their ability to respond to the needs of the patient, indicating that many family members may be hesitant to use drugs to aid sleep.

Impact on social, leisure and daily activities
Family members caring for a relative with a chronic condition experience a considerable impact on their social, leisure and daily activities [38,51,58,65,66], with women reporting greater disruption than men [67]. Most family members caring for their relative reported difficulties in combining caring tasks with daily activities [29,68,69]. Parents of children with chronic disease reported less opportunity for leisure and social activities [38,53,68,70]. The high caregiving demands of children with developmental disabilities, especially if outwardly visible, contributed to social isolation [33]. The parents of children with obsessive compulsive disorder experienced interruptions in social life such as postponing social activities [71]. Parents of children receiving palliative care felt little desire to go out, indicating that the severity of their child's disease led to a loss of interest in leisure activities [72].
There seems to be a cultural aspect to the impact of caregiving on social life. Japanese caregivers reported high social scores on the Zarit burden scale [73], even when their perception of general health was lower than that of the care recipient. This indicates that unlike Western caregivers, Japanese caregivers do not report their feelings about their social life being impacted by caregiving [73]. Arab mothers of children with disabilities experienced reduced social interactions and lower QoL due to the cultural beliefs and the stigma attached to having a child with a disability [48].

Impact on family relationships
A relative's chronic condition has an impact on the relationships among family members and between the patient and the family members [29,74,75]. Caring for a family member not only impacts the carer but also the whole family [16,76] and better family relationships improved QoL for both patient and family members [35,69,77].
Mothers caring for children with attention deficit hyperactivity disorder and oppositional developmental disorder (ODD) experienced negative feelings towards their affected child. Some mothers attributed their child's ODD to increased conflicts between them and their partners [74]. However, having more children was seen as being protective against partner conflict and maternal hostility, as siblings could assist the mother by caring for the sick child, thereby reducing parental stress and negative feelings towards the child [74]. Conversely, siblings may internalise their emotional reactions to the situation, leading to behavioural problems [40]. Better alignment and coordination between parents and involving the siblings, however, could lead to family cohesion, tackling the problem together.
Partners of patients experienced poor sexual life and relationship quality because of the patient's symptoms [68,78], with a significant decrease in the partners' ability to spend quality time with the patient [70], leading to marital conflicts [68]. For many, the caregiving role restricted them from having more children [72]. Knap et al. [72] reported that 48% of parents of children with life-limiting illnesses choose not to have more children because of their child's illness and associated caring responsibilities.

Financial impact
Caring for a relative with a chronic disease can necessitate increased expenditure [15,31,67,68,[79][80][81][82][83]. In an Australian study, the annual personal cost for mild, moderate, and severe atopic dermatitis was calculated at Aus$330, 818, and 1255, respectively, with most being spent on medication, dressings and non-irritant clothing [64]. In a Swedish study, 20% of parents reported experiencing financial difficulties even after the cost of the chronic disease treatment was covered by the welfare system [84]. The family members reduced their working hours or left their jobs to take up their caring responsibilities. This and the expense of hospital visits contributed to their financial difficulties [64,84,85].

Impact on work
Work was seen to have a positive impact on the QoL of mothers, as it provided temporary relief from their caring role, time to socialise and offset the financial burden [47,71]. However, many family members caring for their relative suffered work impairment [75,86] and had to give up their jobs, change jobs, alter career choices or reduce their work hours to look after an ill family member and to manage hospital visits [64,70,87,88].

Positive aspect of caregiving
Despite the physical, social and psychological impact that a relative having a disease has on family members, many family members reported a positive experience of caregiving, with older family members reporting more satisfaction than younger ones [55]. Meriggi et al. [67] reported 93.5% family members caring for their relative were happy with their role. Son et al. [77] attribute positivity in family members caring for cancer patients to their spiritual upliftment. Awadalla et al. [89] attribute this positive impact to family cohesion, and an attitude of hopefulness. Adult siblings caring for their parents reported that they see caregiving as a way of giving something back to parents [90]. Although the health status of family members with caring experience was lower than that of non-carers in an Australian study, the difference in scores did not reach the minimal important difference (MID) magnitude for either the mental or physical domains of SF-12, suggesting that caregivers might be satisfied in their caring roles [91].

Existing family QoL instruments
The appraisal of the family QoL measures identified 48 instruments measuring the impact of a patient's disease on family members. Forty-two of the instruments are disease or speciality specific and are limited to that particular group of patients. The properties of these measures are summarised in Tables 1 and 2.
The review also identified Six population-specific/ generic measures: their properties are summarised in Tables 3 and 4. Five of these measures (Impact on Family Scale, the Beach Centre Family Quality of Life, the PedsQL ™ Family Impact Module, Family Quality of Life Survey and Care-related QoL), are aimed at specific populations of carers (parents of children, family members of people with disability, informal caregivers not necessarily family members of people with long term conditions).    19:194 19:194 19:194 19:194  The only generic instrument that measures the impact of any condition on family members across all specialities is the FROM-16. The HRQoL instruments, regardless of having been developed for patients or their family member/partner, should demonstrate essential psychometric properties such as validity, reliability and responsiveness to change [159,160]. Although most instruments demonstrated good internal consistency, reliability and construct validity, only 11 reported responsiveness and only one reported the MID. Thus, it is not known whether these instruments are sensitive to detecting change over time in family members' QoL.

Discussion
This review has demonstrated that family members caring for relatives with various chronic diseases are impacted in similar ways in terms of physical, social and psychological wellbeing. The high number of FQoL instruments identified in this review demonstrates a growing interest in FQoL, though most research has focused on a few medical fields including neurology, oncology and dermatology, findings consistent with the previous review [5]. One key strength of this current review is that its findings are based on studies that have used valid tools to measure the impact of a patient's chronic disease on a family member/partner. The studies included have used many different instruments to measure the impact of chronic disease on family members, indicating a lack of consensus on the use of instruments: perhaps a clear consensus has not yet emerged because this field is still young. Furthermore, the heterogeneity of the instruments used prevents comparison of the impact of caregiving on family members across disease areas. Such comparison is important in identifying the most vulnerable family members and directing them to appropriate support. This is critical as a physically unhealthy family member would be less able to discharge their caregiving duties, thus having a negative impact on the patients' health [20]. While many studies in this review have used disease-specific instruments, most used generic health status or population instruments to measure the family impact of a person's chronic condition, indicating a strong need for a generic QoL measure specific to family members. Furthermore, most instruments used in this review have been designed keeping patients in mind and may not address issues relevant to family members. Using a measure designed to be family-specific should provide a better understanding of the needs of family members, including support services. Diseasespecific FQoL instruments are used to assess QoL of family members of people with a specific disease and thus can detect changes in family member's QoL following clinical interventions. Generic FQoL instruments on the other hand, can assess the effects of a wide range of diseases or treatment on the QoL of a partner or family member. Published research has shown that family members caring for relatives with different health conditions are impacted in similar ways [161]. Thus, generic FQoL instruments allow the comparison of QoL of individuals across different disease areas and identification of population-wide trends. While disease-specific instruments can help clinicians to understand the extent to which a partner or family member has been affected by a person's disease and inform appropriate treatment decisions, they cannot be used to compare across conditions or between treatments. Moreover, generic instruments can measure the family impact of disease in areas where there are no disease-specific measures. Some research studies may use both generic and disease-specific instruments to capture the different patient/family member viewpoints or to validate the results of using each type of instrument. The FROM-16 could fill this gap as a generic family outcome measure since it has been developed directly from the experience of family members, for family members. One practical feature of FROM-16 is that it is a user-friendly and relatively simple questionnaire with an average completion time of 2-3 min, making it a practical tool for use in a clinical setting.
There are some limitations of this review. The review is not a systematic review. Although not a systematic review, it followed rigorous methodology and fulfilled 19 relevant PRISMA checklist items (Additional file 2: Table S1) [8]. Besides, the review only included studies in the English language, thus limiting understanding of the impact of patients' disease on family members in different cultures. Nevertheless, most studies carried out in different cultures are usually published in English language scientific journals; this suggests the amount of missed information may be minimal. Most studies in this review were cross-sectional. Only five studies were longitudinal, revealing that greater carer burden was associated with poor physical and mental health and lower QoL of family members over time, with women being impacted more than men. Future research should focus on longitudinal studies to build understanding of the long-term family impact of disease. This is important as most acute and chronic diseases may influence major life-changing decisions, thus understanding long-term impacts may help clinicians in developing better management plans for patients and their family members [162]. In addition, the majority of family members caring for relatives in the studies reviewed were women, mostly mothers. There is a dearth of research on the impact of caregiving on fathers, although this review highlighted two studies where fathers were impacted more than mothers. The fact that fathers are mostly unavailable at the point of contact results in the impact on fathers being forgotten or difficult to obtain. Thus, future research should focus on the impact of children's diseases on fathers.
An appraisal of existing FQoL instruments identified a recent plethora of FQoL measures indicating the growing recognition of the importance of FQoL. Only a few instruments have published responsiveness and MID information, however evidence of responsiveness is essential for such questionnaires to be useful for clinical monitoring or as an outcome measure to assess the value of interventions. Information concerning MID is important for the clinician to be able to interpret change in scores over time. Most instruments reviewed were developed recently, and perhaps new studies underway might later report their further psychometric properties. Further psychometric testing of existing measures is required. Furthermore, all instruments identified in this review were created in developed countries, highlighting a need for cross-cultural validation in developing countries [163].

Conclusions
In conclusion, this review found that family members caring for their sick relative experience a huge but similar impact on their physical, social and psychological wellbeing across different disease areas. However, to translate