Quality of life in children and adolescents with Osteogenesis Imperfecta: a qualitative interview based study

Osteogenesis Imperfecta (OI) is a hereditary condition affecting approximately 1 in 20,000 births with eleven recognized types of OI [1, 2]. Children with OI have low bone mass, leading to recurrent fractures, varying degrees of short stature and deformity. Severity ranges from those mildly affected individuals (Type 1) who have minimal bone deformity, near normal stature, blue sclerae and variable hearing loss, to those who are lethally affected with multiple fractures in utero and respiratory failure [2].

Treatment regimes for children with OI aim to; provide pain relief; reduce fractures; prevent deformity, improve mobility and facilitate independent function. The current treatment of choice for children with moderate to severe OI is bisphosphonate therapy. Children treated with bisphosphonate therapy have reportedly shown an increased bone mineral density and reduction in fracture rate [3‐6]. Anecdotal evidence that bisphosphonate treatment increased muscle strength, improved motor function and reduced pain are not supported by controlled trials [7]. Bisphosphonates decrease the efficiency of osteoclasts, the cells which break down bone, and therefore reduce bone resorption [8]. This results in thicker cortices; hence reduced long bone fractures, increased bone mineral density and reduce pain [9]. Thus, OI and its associated treatments can have a large impact on the health, well-being and quality of life (QoL) of the child and their family.

Previous attempts, at measuring QoL in OI, have used several generic instruments including; PEDI, WeeFim, Visual analogue scale, Bleck score, Health Utilities Index III (HUI III), and Self-perception profile for children (SPPC)[10‐12] Although Seikaly et al. [10], using a visual analogue scale and number of pain free days, found significant improvement in wellbeing and reduction in pain between their two treatment groups they failed to identify any improvement in quality of life or function. Generic QoL measures such as the HUI III and the SPPC can be used to assess and compare a range of different disease states and healthy individuals but they may not be responsive enough to detect the small changes in QoL experienced by a child, with a particular disease or condition following treatment and clinically important aspects of the child’s life related to a specific disease may be overlooked [13].

Quality of life (QoL) is a multidimensional concept, which is individual and time dependent, incorporating physical, emotional and social aspects [14, 15]. There is apparent discrepancy across the literature as to what dimensions should be included within a QoL measure, but most agree that it is a subjective measure and should therefore be evaluated by asking the patient.

Disease specific measures may be more sensitive to detect changes in QoL and are useful in comparing different interventions [16] or detecting treatment effects [17], but do not allow comparison with children with different diseases or conditions [15]. We could identify no disease specific QoL measure for children with OI.

Development of quality of life measures is often achieved from expert opinion, literature reviews and interviews with the relevant population [18]. Items are then developed and psychometrically tested on that population. The varied content and different dimensions within instruments is likely due to the developmental process, target population and proposed use [19].

Previously children have been thought to be unreliable respondents of quality of life measures and parents have been asked to provide this information [20]. Parent’s views are often affected by their own experiences of health, their knowledge, life experience and expectations [21]. However, studies suggest the views of children are required when developing new generic and disease specific paediatric quality of life measures to demonstrate good psychometric properties [15, 19, 22]. Children have demonstrated that they do not need their views represented by adults, they are capable of giving insights into their health from 5 years of age [23] and have been used to develop items and a descriptive system for preference based measures [18].

Previous studies [10‐12] within the OI population have not attempted to understand children’s views on their health and quality of life. It is hypothesised that the development of a disease specific QoL measure for children with OI would improve outcome measurement following medical, surgical and therapy intervention. In order to create a disease specific QoL measure for OI children it is necessary to understand how OI impacts on the health and wellbeing of children and their family, and what issues are important to them.

The aim of this paper is to use qualitative interviews with children, parents and health professionals to identify how OI impacts on the quality of life and well being of children and their family. This is the first stage in the development of a disease specific QoL measure for children with OI and aims to identify the concepts and dimensions of QoL that are important to children with OI and their family.

The population was well known to the chief investigator and therefore allowed a hypothesized working conceptual framework, for the health related quality of life outcomes, to be developed (Figure 1) using clinical expert opinion. It was felt the main concept/theme would be the effect of fractures and that this theme may cross cut all others. The intended population was children with OI, aged 6–18, but the potential need for multiple age-related instruments was acknowledged. A frequency based Likert scale was deemed a suitable method for scoring and a three or five point Likert was considered.

Phenomenology was deemed the most appropriate methodology for this study. Phenomenology studies conscious experience as experienced from the first person point of view, and considers how individuals make sense of the world around them [24]. Through observation, attempts are made to gain access to an individual’s common sense thinking, how they interpret the reality within which they live.

Ten children with a diagnosis of OI were purposively sampled from those attending a tertiary Metabolic Bone Disease Clinic in the UK. Participants were approached gradually throughout the study, until saturation was achieved. Saturation of data occurs when no new information is heard or uncovered from the ongoing interviews. The development of QoL measures requires maximum variation within the sample and included children aged 7 to 18 with a range of severity and types of OI. Ten parents of children diagnosed with OI attending the same clinic, but independent of the children sampled; and five health professionals, located within the above mentioned clinic, specialising in the treatment of children with OI were also interviewed. (Table 1 describes the characteristics of the sample). Saturation was monitored continuously throughout recruitment. For completeness we choose to fully recruit to all participant groups to reduce the chance of missed themes.

The local research ethics committee reviewed and approved the study protocol. A letter of invite was sent to participants requesting them to consent to being approached to take part in the study. The invitation included an information leaflet explaining the nature of the interviews, who would be present and what to expect, making it clear that participants could stop the interview at any time. Consent was gained from participating parents and health professionals. For children who agreed to participate, consent was gained from both parents and young adults (16–18 years) and assent was gained from children (7–15 years).

The interviewer [CH] was previously known to the children in a clinical role. This potential conflict of interest was acknowledged prior to ethics committee review, literature and opinions were sought, and following discussion the decision to use a clinician, known to the children, as the primary researcher was made. Children were given the choice to be interviewed alone or with a parent or carer present. Seven children were interviewed without their parent/carer, three chose to have their parent/carer present; these were the three youngest participants. The semi-structured interviews took place in a quiet room, away from the clinic or ward environment. All participants were made aware of the dictaphone recording the interview. Participants were informed that there were no right or wrong answers and that their views and opinions were what was required. An interview schedule was used to encourage ongoing focus throughout the interview. As you would expect of qualitative methodology several warm up questions were used at the onset; parental and child interviews included a warm up question asking them to tell me a little about themselves, health professionals were asked to describe their experience of working with the paediatric OI population.

Following this starter question, they were asked if having OI, or having a child with OI, had an effect on any of their daily activities. Parents and children were asked to discuss their daily routine, we hoped this would allow them to break down their daily activities and explore how OI impacted on their QoL. When interviewing the children probing was used to draw out any differences between themselves and their siblings or peers. Parents who had more than one child were asked if they felt they treated both children similarly.

The interviews were transcribed verbatim, and rechecked on several occasions for accuracy. A sample of interview transcripts (n = 5) were reviewed by a second qualitative researcher [WB]. The interview data was read, and re-read and reoccurring themes were identified. Significant themes and subthemes identified, extracted and organised, utilising framework analysis. Framework analysis is a systematic, visual method of analysing qualitative data and has five stages; familiarisation, identifying a thematic framework, indexing, charting, mapping and interpretation. It is inductive in nature allowing both a priori and emergent concepts to be combined. [25, 26]. Particular attention was paid to identifying differing views in the data. Themes were then explored and any connections or overlap between themes considered. Anonymous data were shared and discussed at research meetings to establish consensus.

The first 25 individuals (10 children, 10 parents and 5 health care professionals), approached agreed to participate in the study. The age range of the children interviewed was 6 to 17 years; the parents interviewed had children with a mixed range of severity of OI. The health professionals included two occupational therapists, one physiotherapist, one specialist nurse and one consultant. All specialised in the treatment of children with OI with a wealth of experience, ranging from 2–16 years. Interviews ranged from 13 minutes to 52 minutes in length (See Table 1).

Most of the children appeared to feel at ease discussing at length their daily routine, identifying times and situations where they needed additional support, or had to do things differently from their siblings or peers. Although one child was able to talk about her Osteogenesis Imperfecta (OI) and describe her daily routine, she became upset when the interview touched on differences between her and her peers so the interview was stopped to avoid causing undue stress.

Health professionals due to their lack of constant proximity to the children and their families often discussed topics from both the child and the parent’s perspective. Children obviously described topics which were more relevant to them alone; parents discussed more family based topics, often with emphasis around the additional planning and organisation involved in caring for a child with OI.

Six main themes were identified as relevant to QoL from the analysis. These themes included; being safe, reduced function, pain, fear, isolation and independence. (See Table 2 for details). Although the complexity of the language used varied, all participant groups discussed the impact of the six themes on the QoL of children with OI, but the emphasis/spread was different for each group. Themes such as; reduced function (fractures and equipment) and isolation (being different) saturated quite early (n = 5), within the children interviewed. Being safe was discussed readily by all children interviewed, but more frequently by parents. The themes fear and being safe saturated earlier within the parent group, and fear was more readily discussed by health professionals. On reviewing the interview transcripts all the main themes were identified within the first eight interviews (4 children, 2 parents, 2 AHPs), although not all the sub themes were identified at this stage.

This study used qualitative interviews with children, parents and health professionals to identify how OI impacts on the quality of life and well-being of children and their family and what issues are important. It was the first stage in the development of a disease specific quality of life measure for children with OI. From the 25 interviews undertaken with children, parents and health professionals, six main themes were identified, extracted and organised using framework analysis and included; being safe and careful, reduced function, pain, fear, isolation and independence.

There is a paucity of research on the views of both children and parents on the impact of OI on QoL.

A recent Portuguese descriptive case study [27] described some similar themes to those identified here. They interviewed children with OI, their siblings and parents. Reporting themes such as; consequences of fracture and impairment, weakness and vulnerability, exclusion at school, worries surrounding pain and susceptibility to fracture and positivity.

Stevens [18] interviewed 74 children aged 7–11, from two city schools, about their quality of life. Her main aim was to develop a health utilities index for children, and used the interviews as a means to identify themes. The majority of children were in good health (93%), no individuals reported complex disability. Her themes included; worried, sad, annoyed, hurt, learning, daily routine, tired, joining in activities, sleep, embarrassed and jealous.

Quality of life is a difficult concept to describe, and descriptions vary across the literature. Previous definitions [14] are based around functional ability and health status, but both experience and the literature tells us that quality of life is not necessarily based on a child’s ability to function, and there is no evidence to suggest this is the case [28]. Difficulties arise when attempting to measure quality of life, if it is not well predefined. Some define it as functional ability or a sense of wellbeing [19], others report health related QoL [29]. It is hard to make comparisons between research papers, if different definitions are used. From the number of themes identified within the interviews, it can be seen that only a small proportion relate to functional ability. Reduced function was more often mentioned if the child had sustained a fracture and this was true for all ages and interviewees.

Several studies have discussed the difference found between children and their parents with regard to their understanding of QoL [15, 16, 30, 31]. This research supports the view that differences also exist between the children and their health professionals [32]. It also highlights a large number of similarities between children and their parents within the six main themes identified, although there were several important disparities apparent within the interviews. Parents talked about how their child’s OI affected their own QoL; their ability to work; undertake family activities; and the additional planning required to achieve some outdoor pursuits. Parents on the whole, failed to mention their child’s independence, without discussing the over protection they felt it necessary to provide. Eiser [16] stated that we shouldn’t expect strong correlations between parent and child ratings, but it is important to identify those contexts in which parents can be expected to make an accurate judgement. Parents are not therefore, necessarily considered an accurate source of information when identifying issues around their child’s QoL [15, 30, 33]. For this reason parents may well be a suitable advocate when reporting on themes such as; reduced function, being safe and careful, and pain, they may not be an adequate proxy overall.

The research undertaken found strong agreement between the groups for themes; being safe and careful, functional ability particularly following fracture, isolation from activity and pain. Although all three groups agreed on the nature of these themes, their justification for the theme was different. There was not agreement however when discussing independence and fear. Young children of all abilities are cared for by their families, so striving for independence is often not anticipated in this age group. With the exception of the very young children, fear was a theme reported by all those interviewed. The impact of fear varied across the group; children had a fear of certain activities that previously resulted in a fracture, parents fear related to anxiety of handling and of safety/separation when the child was at school/nursery or play dates. As expected younger children were unaware of the over protective actions of their parents which is a stark contrast to the older children expressing their dislike of parents attempts to keep them safe and doing everything for them.

As anticipated many of the themes have a link to fractures; the fear surrounding potential fracture, the avoidance of sustaining a fracture, the resultant pain and reduced function following fracture, and the effect of repeating an activity which previously lead to a fracture. No other literature or QoL measure contains any link to fractures and the effect they can have on a child’s QoL. Several of the themes identified from the interviews are similar to those in other QoL measures. Independence, function and pain are included within the CHIP-AE [34], TACQOL [35], and DISABKIDS-37 [36], but this lack of connection to fractures demonstrates the need for an OI specific QoL measure.

This data presents the first step in developing items for a disease specific QoL measure for children with OI. Six main themes were identified; being safe and careful, reduced function, pain, fear, isolation and independence. There was generally good agreement between the three groups of interviewees, although discrepancies did occur between parents and children, with regard to the themes independence and fear. Several of the themes uncovered showed similarity to other QoL measures, but the addition of being safe and careful, particularly in relation to fractures, demonstrated the need for a disease specific measure for children with OI. Further research in the form of focus groups will take place to validate these themes. The themes identified and vocabulary used by the children will directly inform development of the items within the OI specific QoL measure.

CH (MSc, PGCE, MCSP) is a practicing paediatric physiotherapist within a specialist Metabolic Bone Disease Service and undertook this research as part of a PhD qualification.