Listening to the experts: Parents' perspectives around infection risk and returning to education and social activities following their child's diagnosis of acute lymphoblastic leukemia

Abstract Background During a child's prolonged treatment for acute lymphoblastic leukemia (ALL), there is a need to balance their increased risk of developing infection‐related complications with meeting their educational and social needs. Aims To determine the safe timing of return to social activities for children undergoing treatment for ALL and to determine how parents perceive and act on advice related to infection risk while navigating their child's “return to normal.” Methods and results Medical and educational attendance records were reviewed for 47 children who were diagnosed with ALL and 24 semi‐structured qualitative interviews were conducted with a representative sample of their parents. The majority of children (69%) did not return to education prior to the start of maintenance therapy regardless of the advice that the families received from their healthcare team. Those who returned earlier were at no greater risk of major infection complications (mean = 0.5) than those who did not return until after commencing maintenance (mean = 0.4, P = .74). Parents spoke of the difficulty in obtaining practical, consistent, and timely advice and of balancing infection risk with a desire to return to normalcy. Inconsistent advice and constant vigilance placed a burden on parents which often profoundly affected their mental wellbeing. Overall, parents wanted to make their own decisions about how and when their child returned to education and social activities. They made these decisions based on many factors, of which infection risk was just one. Conclusion Following the study conclusion, a national working group was established—including parent representatives—to implement the study recommendations. This includes the development of a range of practical resources to better support families. Health professional guidelines provide quantitative data pertaining to infection risk, while emphasizing that the returning decisions ultimately rest with the families. This research demonstrates that listening to parents—who are the experts through their lived experiences—is a critical element in creating policies that are responsive, meaningful, and widely accepted.


| INTRODUCTION
Acute lymphoblastic leukemia (ALL) is a disease of the blood-forming tissues of the bone marrow, which can occur at all ages, though it is primarily a disease of childhood. 1 ALL accounts for over one in four of all new cases of cancer in New Zealand children each year and currently has a 5-year survival rate of over 90%. 2 The long course of immunosuppressive chemotherapy treatment-typically 2-3 yearsmeans that there is an extended period of risk of infection-related complications, which can lead to changes in treatment, deferral or interruption of a treatment phase, and re-hospitalisation. [3][4][5] Infections occur across the whole treatment period, with some infections more common during the induction phase, 3,6 but higher numbers of infections documented during the longer maintenance phase. 5 Infection-related complications can lead to costly additional hospital admissions, resulting in further stress and upheaval for the child and their family.
For parents, caregivers and health professionals, there is a desire to find a balance between the risks of infection for children undergoing treatment for ALL and meeting the child's social and educational needs. The return to social activity parent-directed resources and the academic literature have largely focused on why and how, 7-10 with the implication that returning to education and social activities should happen as soon as possible due to the psychosocial benefits for the child as well as the wider family considerations such as parents' ability to return to paid work. 11,12 Despite these recognized benefits, medical professionals-even within the same workplace-have been shown to hold differing opinions about infection risk and the timing of return to social activities. 13,14 There is currently no existing body of research that has determined an optimal approach with regard to when it is safe for children undergoing treatment for cancer to return to education and other social activities. 14 All children diagnosed with cancer in New Zealand have their care coordinated by a specialist multidisciplinary child cancer team based at either the Children's Haematology Oncology Centre (CHOC) in Christchurch or Starship Blood and Cancer Centre (SBCC) in Auckland. These two specialist pediatric oncology treatment centers work closely with 14 shared-care pediatric teams so that patients can receive as much of their care as close to home as is safely possible.
The child's induction phase of treatment is undertaken at one of the two specialist centers, while the remainder of their treatment can predominantly be administered at the child's local shared care hospital, with additional scheduled appointments at their specialist center.
There appeared to be a difference in the standard advice provided by SBCC and CHOC regarding the safe timing (related to infection risk) of return to social activities and education following an ALL diagnosis. Anecdotal evidence suggested that CHOC families were advised that their child could return immediately after induction therapy while SBCC families were advised to wait until the start of maintenance treatment. Figure 1 illustrates that for children being treated according to a sample standard risk B cell ALL protocol this potentially meant that there was a 28 week period in which a CHOC patient was attending school or early childhood education (ECE) while their counterpart at SBCC was not.
The differences in standard advice between New Zealand's two specialist centers provided the opportunity to address a gap in the international literature regarding the safe timing for children undergoing treatment for ALL to return to education and social activities and to use the results of the study to inform nationally consistent guidelines for New Zealand's two specialist centers. Initially, our study was conceptualized as a simple retrospective review of patient medical records to determine if those treated at CHOC-who were advised to return to education at the end of induction-had a greater number of infections that those treated at SBCC-who were advised to return to education at the start of maintenance. However, at the study design phase it became evident that a medical record review alone would not address firstly, if the many individuals who provided infection control and return to social activities advice to patient families were consistent in the advice that they provided and secondly, if families followed the advice that they received. That is, did children return at the time that families were advised that it was medically safe for them to do so? The study was therefore expanded to become a mixed methods study which incorporated an online survey of healthcare and other professionals who advise families about infection risk and returning to school activities; a retrospective review of both the medical and educational attendance records for children diagnosed with ALL in each center; and in-depth qualitative interviews with a representative sample of the children's families. These three study components were undertaken concurrently.
Here, we focus on two of the study objectives; whether returning to education prior to the start of maintenance therapy was associated with higher rates of infection in children with ALL; and how families Quota sampling ensured that those families who were selected to be interviewed were representative of the total cohort in terms of their child's age at diagnosis, treatment center, sex, and ethnicity (see Table 1).
For the medical and educational records review participation rates were higher for families treated by CHOC (24 out of 39 families contacted; 62%) compared to those treated in New Zealand's larger specialist center at SBCC (23 out of 58 families contacted; 39%). In addition, participation rates were lower for Maori and Pacific families, although this was able to be addressed through quota sampling for the qualitative component of the study. Aside from this, the study participants were largely representative of the total cohort and similar across the two centers according to a range of characteristics including sex, involvement with a shared-care center, enrolment in a clinical trial, risk group, level of deprivation, and family size.
The median age of the study participants at the time of their diagnosis was 3 years and 9 months, reflecting the age at which the incidence of ALL peaks. 1   Following the retrieval of data from the educational and medical records, all variables were coded and merged onto a single spreadsheet. As the period between the end of induction therapy and the start of maintenance was when the return-to-school advice potentially differed between the two centers, this was the specific timeperiod that was investigated. Data were analyzed to determine whether the infection rates at the specialist centers differed and whether return to education prior to the start of maintenance was associated with a greater risk of infections. Planned comparisons were conducted to compare the rate of "early" return to school by specialist center and the average number of half days that children attended between the induction therapy and start of maintenance. All statistical analysis was conducted using Stata version 14. A deductive thematic analysis approach was used to code relevant interview data chunks. Following the initial coding of 81 descriptive categories, thematic networks were used to determine how they interrelated and to create 14 initial themes. 18 A second member of the research team reviewed a sample of five of the 24 interviews (21%) to assess inter-rater reliability. [19][20][21] This showed 83% alignment across 243 identified themes from the sample interviews, which is well above the 70% considered "acceptable." 19 Together the two researchers undertook a process of inter-coder agreement by reworking the categories and themes in collaboration to reach consensus. 19 The final result was 79 categories and 11 themes, which were further distilled down to three overarching themes: consistency, balance, and mental health and wellbeing.

| Medical records review
The overall number of infections recorded between the end of induction and start of maintenance was low, with 64% not developing a single infection that resulted in a deviation from their treatment plan during the study period. The mean per patient infection rate was 0.4 for CHOC patients and 0.7 for SBCC (P = .48, see Table 2). Those children who returned to education prior to the start of maintenance therapy were at no greater risk of infection (mean = 0.5) than those who did not return until the start of maintenance (mean = 0.4, P = .74). Treatment according to a high-risk protocol was the only variable examined which was associated with a higher rate of infections (mean = 1.2, P = <.01).

| Educational attendance record review
This study was based on the belief that children undergoing treatment for ALL at CHOC were being advised to return to education after induction therapy while children treated at SBCC were being advised to wait until the start of maintenance. We would therefore have expected that most children treated at CHOC would have returned to school/ECE prior to maintenance therapy and conversely very few of the SBCC patients would have. However, an analysis of the educational attendance records showed there was no statistically significant difference between the two groups in the time period under investigation. Nearly one in four children (23%) who were treated at SBCC did return to school/ECE at some point prior to maintenance and although the return figure was higher for children treated on CHOC (40%, P = .23), this still means that the majority (60%) of CHOC patients did not return to school/ECE at all during this period.
The mean number of half days attended for SBCC patients was 2.1 days compared to CHOC patients was 17.6 days (P = .12, see Table 3). This non-statistically significant difference was largely due to two clear outliers. One CHOC child attended school for 88 half days during this period, while another attended ECE for 152 half days.

| Parent interviews
Twenty-four parent interviews were conducted using a semistructured interview schedule. The parents, who were interviewed, were a representative sample of the entire ALL cohort in terms of their child's gender, ethnicity, age at diagnosis, and treatment location. As the cohort included children diagnosed across a 2-year period, the parents interviewed were all at different stages of treatment, ranging from maintenance therapy to 1-year off treatment.  Without consistency parents sought other opinions, felt a loss of control, and started to doubt themselves and their decisions. Inconsistency could increase a parent's concern that they might not be being given the best advice, that they have made a poor decision or put their child in danger.
And because they got slightly different advice we did think oh, maybe we're taking her back a bit too much.
( In New Zealand, children whose health prevents them from regularly attending primary school are eligible for admission to a Regional Health School. It was notable that while parents often spoke positively about the health schools, they did not consider their child's attendance as signifying a return to social or educational activities.
And I think it's that whole normalisation. Trying to make a really abnormal situation as normal as possible.
And school was a normality for us… Northern Health School were fantastic but it's still not the same as being at school. (Interview 17)

| Mental health and wellbeing
Although it may be expected that having a child undergoing cancer treatment would impact a parent's mental health and wellbeing, the interviews illuminated many areas where parents could have been Educating family, friends, and others about their child's infection risk was often an additional stressor for parents. Parents expressed a desire for practical resources so that they were not burdened with educating others at a time that they were already physically and emotionally exhausted.
When someone is going back to school or kindy, for there to be like a template or something that's really easy to access to give to the school or kindy.

| DISCUSSION
The medical records review was conducted to address a fundamental gap in the literature on post-diagnosis experience for families with ALL, namely that there is no empirical evidence to inform the timing of return to education for ALL families. 14 It was based on anecdotal evidence that different advice was being provided at New Zealand's two specialist centers which afforded us the unique opportunity to undertake a "natural experiment" to determine if there was an increased risk of infection for CHOC patients-who returned to education after induction-compared to SBCC patients who did not return until the start of maintenance. If there was no difference in infection rates between the two groups, then we could confidently advise all New Zealand families with a child undergoing treatment for ALL that it was safe for them to return to education from the end of induction, with the presumptive educational, social, and family benefits which were promoted by this earlier return. 11,12 Unfortunately, this study not able to conclusively answer the research question on the safe timing of return to education. This is because 60% of CHOC patients did not return to education until the start of maintenance, irrespective of the advice that families may have received from their child's oncologist and others who acted in an advisory role. Our study did, however, find that there was no increased risk of infections for those children who did return to education between the end of induction and the start of maintenance therapy.
In addition, the number of infections recorded which resulted in changes to the child's treatment plan were low. In fact, two-thirds of the children in this cohort did not develop a serious infection according to our study criteria. We hope that this finding can be of some reassurance to those families who are concerned about their child developing an infection once they begin to interact more regularly with others outside of the hospital environment.
The reasons why we were unable to answer our first research question and why the group of children receiving treatment for ALL through CHOC did not return to education any earlier than those from SBCC became apparent through the qualitative component of the study. The 24 parental interviews highlighted two major erroneous assumptions that we had made when we had first conceptualized this piece of research. Firstly, we believed that although the advice around infection control and returning to social activities given to families at each the two specialist centers differed, each family was receiving largely consistent advice from the medical and allied health professionals involved in their child's care. Yet through speaking directly to the families, it became evident that the advice that they were being given was not consistent even from staff at their child's specialist center, let alone between the multitude of others who gave advice to the family over the course of their child's treatment. The inconsistent advice that many families received and the significant negative impact that this had on parental mental wellbeing was a key theme to emerge from the interviews.
Secondly, we wrongly assumed that infection risk was the key consideration when parents made decisions around returning to social activities and education and that parents would be keen for the child to return to social activities as soon as it was declared medically safe for them to do so. However, our parental interviews revealed that infection risk was but one of many factors that these parents considered when navigating their child's "return to normal." Parents did not make decisions about infection risk and returning to education in isolation, or just with a health focus, but balanced these decisions within the wider context of their lives and their family. Such decisions required them to weigh up financial, emotional, and practical considerations. Their child's personality, response to treatment, and the needs of other family members all featured heavily in their decision making.
Through the parental interviews, it was clear that parents want to make their own decisions about the timing of their child's return to education. Parents make these decisions by carefully considering their child's needs within the context of the needs of the wider family. [22][23][24] However, to make such decisions an empowering process for the parent, they require clear and consistent information delivered in a way that is tailored to meet the unique needs of their family. [25][26] Without this, parental perceptions of the health team's capabilities can be reduced. 27 The inconsistency of advice, the need to educate others and the burden of responsibility were articulated as areas of additional pressure on parents, particularly at a time when parents were already physically and emotionally stretched. This decisional burden and the emotional work in caring for a child with a major health issue can negatively impact parental mental health. [28][29][30][31][32] An important way for parents to feel a sense of agency is to provide them with information and advice which allows them to make informed decisions. 24 Our research indicated a clear role here for cancer health and support services to support parents' and children's mental health and wellbeing through providing clear and consistent advice to families. Ultimately, consistent advice and information will help parents trust in themselves and F I G U R E 2 Study recommendations the decisions they make while also giving them greater trust in their health team. 31,32 Following the study conclusion, five major recommendations were made to ensure that the key study findings were addressed (see Figure 2). The National Child Cancer Network established a working group-which includes medical, NGO, educational, and family representatives-to implement the study recommendations. This is the first time that parents have been included in a National Child Cancer Network working group and emphasizes the important role that families can play in ensuring the research findings are successfully translated into practice. The working group will develop a range of practical, user-friendly resources both for the parents themselves and for them to share with friends, family, school, and their wider social circle. These resources will include direct "parent to parent" advice, which incorporates quotes taken directly from the parent interviews.
The new health professional guidelines will provide quantitative data pertaining to infection risk, while emphasizing that the decisions about when and how a child returns to education and social activities rest with the patient families. All health professionals and others who advise families about infection risk and/or returning to education and social activities will be provided with these guidelines to help ensure that the future cancer families receive consistent advice and support during their "return to normal." Our original study was conceptualized as a single analysis of infections recorded on patient medical records to identify when a child undergoing treatment for ALL could safely return to education and social activities. Yet had we relied solely on the medical records, we would have concluded that it was safe for a child to return after induction therapy but have failed to identify that many children treated on CHOC were not in fact returning to education at this time.
We would also not have gained the insight that infection risk was but one of parents' many considerations when navigating their child's "return to normal." Ultimately, this would have led to the development of national guidelines that were not fit for purpose or widely adopted. Our study illustrates the importance of listening to parents if we are to create policies that are responsive, relevant, and well received.
Although this paper has discussed the findings from research with a specific population-ALL patients and families from New Zealandwe believe that our learnings around the importance of consistency of advice and supporting parental wellbeing have a much wider application across a variety of populations and contexts within child health.
Going beyond the perspectives of healthcare providers to learn from parents' lived experiences will add to current bodies of knowledge and potentially offer new opportunities to identify solutions to improve clinical practice. The parents that we interviewed welcomed the opportunity to participate in research and contribute to initiatives that had the potential to improve the experiences of the families that come after them. This concluding quote reflects a commonly expressed sentiment: If there is anything that we can do to provide information and help out that can improve things for new patients [and] new parents in any way-even one person-it's good.