Preterm children quality of life evaluation: a qualitative study to approach physicians’ perception

The consequences for extremely preterm children (EPC) (<;28 weeks’ gestation) are well known in terms of mortality and morbidity[1, 2]. Interest is increasing in not only these traditional health indicators but also in the subjective representation of health and well-being, also known as health-related quality of life (HRQoL); the subjective evaluation of quality of life (QoL) is possibly as important for the patient as the traditional focus on morbidity and the prolongation of life[3, 4]. HRQoL evaluation seems important because complications affect the daily basis of formerly EPC during early school age[5‐7]. However, even if data on HRQoL for formerly EPC are available, they are limited because age groups are limited; indeed, if consequences on HRQoL seem more marked at early preschool age[8] than at the teenager[9] or the young adult[10, 11], this report can be biased because of the mode of evaluation of the QoL which varies with the age of the subject and according to the methodology[3, 4].

In chronic diseases, the perceptions of physicians of the HRQoL of their patients often differ from those of the patients and their families[8, 12‐14]. While HRQoL data are available, there are little data on the basic understanding and perception of quality of life issues by perinatologist physicians as well as the day-to-day experience of the patient or the role it plays in clinical decision-making[15‐17].The need for quality data in this area is all the more important as these physicians are directly involved in the decisions taken with regard to neonatal resuscitation (even if these decisions take into account the families’ opinions) as well giving us a more stable platform for relaunching the public debate. What is the impact of the neonatal resuscitation in term of quality of life?

It would be relevant to highlight the opinions of perinatology experts regarding EPC HRQoL, and in particular physicians because they are the decision-makers. We report in a preliminary qualitative study the first step of the project. The aim was to produce themes of reflection and identify subjects that would imply study on a larger scale.

Eligible participants were experts who manage EPC at the French University Hospital of Marseille, composed of four hospitals. Two of them are involved in the care of very preterm children (level 3 maternities), one in the long term follow up of the children; another hospital is involved in the long term follow up. The participants were defined as follows: more than 18 years old; both genders; having a minimal level of resident practice; involved in the care of the EPC (obstetricians, intensive care physicians, neonatal physicians, psychologists and paediatric neurologists); having agreed to participate in the study; registered on the list of the staff working in the University Hospital. The participants were selected by drawing lots stratified by the main clinical functions from the list. They were asked to complete a short socio-demographic and clinical characteristics questionnaire.

We used the qualitative method of focus groups (FGs) to generate data using the opinions collectively expressed by the participants[18‐20], as it has been used in other contexts and more specifically with children[21].

Focus groups were designed to include at least one representative from each specialty. Confounding factors such as age, gender, function and practice were also considered; according to the hospitals, preterm children can be cared in departments of neonatal intensive care units where neonatologists work, and in mixed paediatric and neonatal intensive care departments (newborns and older children) where intensive care physicians work. The FG were organized according to the availability of the participants. A minimum of two FGs was planned, but the number was increased until a saturation of information was reached beyond which no new concepts were emerging from the FGs. Five new participants would be selected from new drawing lots.

Focus groups were audio-recorded and moderated by a well trained social psychologist (SC) using a discussion guide with guidelines and open-ended questions. The guide was based on an analysis of the literature and centred on the prioritised themes to be discussed: HRQoL representation, expectations in daily practice, HRQoL evaluation (i.e. Table1). The psychologist supplemented the prepared questions with sub-questions that enabled the researchers to explore participants’ answers in greater detail. She encouraged equal participation from the whole group and encouraged people to engage with one another to create a dynamic discussion that would characterize the specificity of group conversations rather than individual interviews.

All the sessions were manually content-analysed by the social psychologist. The data were coded and classified by reviewing the transcribed discussions for categories, using the guideline questions as initial categories. Data relevant to the research questions were identified. Within categories, the different items were written up as statements. Relevant statements that did not fit into the predefined categories were assigned to new categories[19, 20, 22, 23]. A second independent review of the data codes and categories was made by a physician (MAE). At the end, the coherence of this content analysis was confirmed by an external expert from the Self Perceived Health Assessment Research Unit (MCS). Opinions were compared between the participants’ characteristics (gender, age, experience, specialty and practice-type).

All participants provided informed consent. Anonymity and confidentiality of information were guaranteed. Ethical approval was obtained from French legal authority: the National Committee of Computing and Liberties.

In this preliminary study, focus groups made appear themes of reflection classically told about HRQoL in other domains but not all in so clear a way (definition, interests and limits, ethical reflection)[13, 14, 24, 25].

All the themes discussed hold considerable interest for the participants. These themes touched many dimensions: QoL as both a theoretical and practical issue, the physicians’ practical experience, that of the value as well as the limits to this type of approach. However, three sub-themes seemed particularly relevant to the group.

The first one of them concerns the definition of the quality of life.

The FGs’ participants seemed to have embraced the definition of HRQoL given by Eiser and Morse, identifying some key elements such as “subjectivity and multidimensional aspects”[26]. Although they recognized a subjective aspect, the participants indicated that a parental approach was necessary. This pragmatic attitude is in accordance with several studies in which parental evaluation was realized, especially with children less than six years old[5, 6, 8, 27].

The second sub-theme concerns the limits of the use of quality of life concepts.

One of the limitations identified by users is that it seems difficult that the construct underlying the QoL is not constant at all ages of life of children and adolescents. The adaptive process relative to these specific periods of life are multiple referring indifferently to complex process of redefinition but also of recalibration of response or reprioritization of domains. The nature of HRQoL must be renegotiated throughout life[28]. Therefore, it seems illusory to imagine that what structures the QoL of each individual is invariable. For the participants, this evolution would prevent any standardized HRQoL evaluation. Because a child is continuously maturing, developing and changing, then the whole concept of HRQoL is invalid anyway.

Why would this “evolutionary” factor be a limitation for subjective measurement and not for objective measurement? One obstacle to using QoL measurement seems to be related to the implicit comparison with a standard or norm (perhaps because of a misunderstanding about the tools of measurement). Obviously, this subject is often reported on by experts who argue for the subjectivity of the measure used. However, in the field of so-called “non-objective” measures, many indicators are used to assess consensual aspects, such as intelligence quotient. Additionally, one might similarly question the value of several of our so-called objective measures, such as neurocognitive assessments, behaviour disorders, children’ size… The standards of certain objective criteria are not without problems. This issue raises the problem of the definition of the standard and its utilization. Canguilhem discussed the individualization of the standard in his criticism of the positivist determination of the normal and the pathological[29]. The “pathological” becomes an experience lived before being measured. Mistrust towards the objectivity of standards is, henceforth, a common problem in contemporary ethical culture. Using HRQoL standards does not mean defining a “normal life” but improving the everyday life of patients according to their situations[30, 31].

Finally, the third sub-theme concerns decision-making with respect to QoL.

The main question addressed in the FGs was whether HRQoL assessments can assist in evidence-based decision-making. HRQoL seems to have a limited impact on perinatal guidelines. Globally, HRQoL is almost exclusively used as secondary criterion[17, 32, 33]. What would be the limit of this criterion if HRQoL data were available? There are many paradoxes. Participants said they were not familiar with HRQoL tools, but they knew the different aspects of the debate surrounding the topic with real practice attitudes. The differences in the evocation of QoL concept among “ethical thinking staffs” in perinatal units for withdrawing therapy (a way of clearing themselves of responsibility by taking into account at least one time what people could feel?), the lack of knowledge concerning evaluation methods and the fears expressed in the FGs do not position HRQoL as a criterion for making decisions in perinatal situations.

There were several limitations to our study. Participants were selected from one geographic region, and their experiences and opinions may not be generalizable. It is recognized in FG research that the recruited sample is not representative of the entire population but is rather a snapshot of those people participating in the study[34]. The presence of dominant participants and opinion leaders could have been a confounding factor; the experienced social psychologist controlled this issue. Some of the differences in perspectives that have appeared to be conflicts of opinions, can be explained by the varying professional and social positions of the participants. The sample size of the FGs should be discussed: to optimize group discussion, the size of the groups must be limited to avoid the creation of sub groups; at the same time, the groups must be large enough to allow for the emergence of a variety of controversial themes. The sample size and the limited number of FGs could appear as a limitation but saturation of information was nevertheless reached. Most authors support that adequate size of each group is between four and twelve individuals and the use of a minimum of three FGs with each type of participant; sufficient FGs have been conducted when no new information emerges from the dialogue of subsequent groups[34‐36]. Focus Groups’ composition was not homogeneous because they were organized according to the availability of the participants. The heterogeneity of the participants allowed for the expression of different opinions about the subject. Paediatrician neurologists were included with perinatalology experts because they have experience in specialized EPC follow-up care. The participants were mostly female but it is a representation of the French medical reality and especially the profession of paediatrician, as it is described in the French review of social fields by Lapeyre and Le Feuvre [Unpublished observation: “Feminization of the medical profession and professional dynamics in the field of the health” Revue française des affaires sociales, 2005, 1: 59–81] and in the report of the National Center of health professions’ demography[37].

Focus groups are used to explore topics on which little research has been conducted and have the advantage of enabling researchers to quickly identify the full range of perspectives held by participants[35]. Moreover, the interactional nature of FGs allows participants to clarify or expand on their contributions during the discussion in light of the points raised by the other participants.

Our original, qualitative study provides preliminary results about physicians’ perspective on the health-related quality of life of extremely preterm children. This approach was never developed in the field of prematurity with well-knowed consequences on QoL. These results require to be confirmed on a larger representative sample including leaders of obstetrics, neonatal medicine, neonatal intensive care units of French University Hospitals as well french paediatrician neurologists. The themes and questions of this broad opinion survey will rest on the information issued from our preliminary interviews.