Parents’ information needs, self-efficacy and influences on consulting for childhood respiratory tract infections: a qualitative study

Between October 2010 and May 2011, seven focus groups and 30 interviews were conducted. The focus groups aimed to investigate parents’ pre-consultation beliefs and behaviours and these parents were recruited from the community. The interviews aimed to investigate parents’ views and experiences following recent consultations for children with cough recruited through primary care practices. The focus groups each comprised 4-6 mothers and were recruited from pre-existing groups, including parent and toddlers, ante-natal class friendships, and groups associated with schools and community organisations. Parents with children between 3 months and 12 years were invited to take part. Those who only had children less than 3 months of age were excluded because they can have different clinical features (e.g. less likely to develop fever) and are at greater risk for serious complications than older children [17]. Parents with older children were excluded because they are less likely to consult [17] and older children may play an increasing role in decision making, whereas this study was concerned with parental decision making. Purposive sampling was used to select participants to capture maximum variation in views and experiences. The groups were stratified in relation to parents’ socio-economic situation (SES) and age of children, and only mothers volunteered for the focus group discussions.

Eligible parents for the interviews were identified through a search of patient records, in six GP practices, for those who had consulted in the previous 3 months for a child with a respiratory infection; where more than 60 eligible parents were identified by a practice, 30 letters of invitation were sent to the parents of children who had consulted the doctor (for any cause) most frequently during the past 12 months and 30 letters were sent to parents of children who had consulted least frequently during the same period in order to include parents with a range of consulting frequencies. Twenty-one mothers and two fathers volunteered and were interviewed. In order to increase the socio-economic diversity of parents in the interview sample, 7 younger mothers who had participated in the focus groups were also recruited for interviews.

Focus groups were conducted in a range of non-clinical community settings. Most interviews took place at patient’s homes with a few in other non-clinical settings. All participants received both written and verbal information about the research and provided written informed consent before the group or interview. The focus groups were facilitated by three researchers, one led the discussion using open-ended questioning techniques to elicit participants’ own experiences and views, and ensure all participants had an opportunity to take part; one summarised the discussion on a flip chart to allow for group checking and reflection during the session; and the third audio-recorded the session and noted down who was speaking to aid transcription. Each focus group lasted between 60-85 minutes. Interviews were conducted by one researcher and lasted between 30 and 90 minutes. There were separate topic guides for the focus groups and for the interviews. The topic guides for the focus group investigated parents’ pre-consultation beliefs and behaviours. Topic guides for the interviews investigated parents’ perceptions and experiences of primary care consultations when their child had a cough. Topic guides, informed by the Health Belief Model, were used to facilitate the focus group discussions and semi-structured interviews (See Table 1).

Focus group data collection and analysis were conducted in parallel and groups continued until data saturation was reached and no new themes were arising from the data [18]. Preliminary findings and questions from early focus groups were raised at later groups for further discussion. Findings from the focus groups helped to influence the topic guide for the interviews which also continued until data saturation was reached. Focus groups and interviews were audio-recorded, fully transcribed, anonymised, checked for accuracy and then imported into the software package NVivo8. Thematic analysis [19] , using the constant comparison technique [20], was used to scrutinise the data to identify and analyse patterns across the dataset. Transcripts were examined on a line-by-line basis with codes being assigned to segments of the data to provide insight into the participants’ views and beliefs and an initial coding frame developed. The process of constant comparison allowed for the coding frame to be modified as analysis developed, and allowed us to generate new themes, re-classify existing themes and incorporate them within others. Two focus groups and two interviews were coded independently by two researchers; any discrepancies were discussed within the research team to achieve a coding consensus and maximise rigour. The index of multiple deprivation (IMD) score for the parents’ home post-codes was obtained and they were categorised into IMD quintiles. These quintiles were then used to group parents into those living in high-SES (affluent) areas (two least deprived IMD quintiles), mid-SES areas (middle IMD quintile) and low-SES (deprived) areas (two most deprived IMD quintiles) and a comparison of themes by SES was carried out.

In total 60 parents took part in the study; 30 in the focus groups only, 23 in interviews only and 7 in both. Table 2 describes the wide range of socio-economic areas, education level and age of the parents who took part. Most families had 1 or 2 children ranging in age from 5 months to 17 years old; most parents were of white-British ethnicity, 2 parents were white-other ethnicities and 5 were of non-white ethnicities (3 Black, 2 Asian, 1 mixed).

Themes developed from the analysis related to the Health Belief Model were: perceptions of personal threat from illness; the perceived benefits of, triggers and barriers to consulting for a cough; and perceived parental efficacy as shown in Figure 1. Quotes from parents illustrate the points being made and they are identified by high, mid or low socio-economic status (SES) and ages of their children.

Experience was the key factor which parents reported increased their self-efficacy [21], that is their confidence that home care was likely to be successful and thus reduce their need to consult or re-consult the doctor. Less experienced parents described their difficulties in differentiating between serious and minor coughs and some admitted to finding the frequency with which young children suffered from cough a surprise – highlighting the role of un-realistic expectations and the need for more appropriate information.

Parents described how they were unable to obtain helpful information or advice to facilitate their decision to consult. In relation to cough, they expressed uncertainty about the way to distinguish between a “normal” cough which would be self-limiting and a cough which needed to be seen and treated by a doctor. They were then consulting because they were not sufficiently reassured by the information they had found and felt it safer to consult the doctor.

“I don’t want a prescription every time I go, I just want reassurance that I’ve done the right thing in coming.” (high-SES, 2 children 7m & 23m)

Views were compared and across and between groups of different socio-economic status and with different ages of children. Although some differences emerged within groups, they did not differ substantially between the groups.

Our study found that parents from all socio-economic backgrounds sought information from a wide range of sources about RTIs in children in order to identify which of their child’s symptoms they should worry about and trigger a visit to the doctor. The personal threat of a cough to a child as perceived by parents included a combination of the severity of the illness and the susceptibility of a particular child to developing a cough (as predicted by the health belief model). Information was also sought to support self-care and increase parental self-efficacy to care for their child at home. Parents sought specific advice about their child’s current circumstances, rather than general advice about when to consult. The role of friends and family as important sources of such information for parents across all groups was highlighted. Experience with other children increased perceived self-efficacy. Interestingly for our understanding of triggers to reconsultation (when parents return within the same illness episode), the consultation itself was seen to be beneficial regardless of treatment decisions, because it secured a medical evaluation, reassuring parents and providing them with knowledge. Relating our findings to the psychological models of health behaviour has helped to understand help seeking behaviour and reconsultations.

Kai [2, 6] (exploring acute childhood illness) and Francis [22] (for RTIs) included only parents from deprived populations in their studies, since they are known to be higher consulters in primary care, but our study shows that parents from all socio-demographic groups have similar concerns and needs. Kai’s study showed that disadvantaged inner city parents reported feeling under pressure when their child was ill and this was linked to their sense of personal control and the perceived threat of the illness [2, 6]. Francis’s study [22] reported that parents commonly express a desire to have greater information about recognising serious illness in their children and his more recent study highlighted the difficulties that parents have in interpreting the importance of signs and symptoms of serious RTI which may cause delays in consulting [23]. Neill [24] also explored acute childhood illness within family life and similarly reported parents’ uncertainty about when to seek medical help; also confirmed in our study. Studies exploring internet use for health information have emphasised the need to guide parents regarding reliable resources online [25] and reported parental preferences for online clinical health information to be presented by clinical professionals, and online parenting advice to be presented by other parents [15].

Influences on parental help-seeking behaviour have been explored in common childhood problems in toddlers [26] and for child mental health concerns [27]. Consulting was linked to parental perception of illness severity [5], child’s susceptibility to the illness, perceived parental efficacy and ability to cope, their knowledge of the illness and the availability of services. Others have used health belief models to explain the psychological factors influencing help seeking consulting behaviour [10, 11], or have modified it to reflect a health promotion stance for young families [28]. This latter approach reoriented the health belief model and based it on positive health definitions associated with health promotion.

Parental focus on the specific features of this illness episode (susceptibility to this type of illness, severity of this episode) could explain why the health belief model has been less successful in explaining take up of routine care [29]. Our study illustrates that a combination of: the perception of the severity of this illness; together with social and factual information sources; and parental efficacy, informs parents’ decisions about whether to consult the doctor.

Ours is the first study to explore these issues in the context of acute childhood RTI with a range of parents from different socioeconomic backgrounds, enabling a complete picture of views. Although some differences emerged within groups, they did not differ substantially between the groups. Findings from the focus groups were then explored further in the interviews particularly around illness susceptibility and severity and the costs and benefits of consulting. Views were almost exclusively those of mothers since only two fathers attended for interview and none participated in the focus groups, so this study was unable to make meaningful comparisons between the experiences of mothers and fathers. The data analysis was robust as coding was independent, attention was given to contradictory views and themes arising were discussed within the research group.

Ethical approval from Southmead Local Research Ethics Committee 10/H0102/55.