Article Text

Download PDFPDF

Social circumstances and medical complications in children with intestinal failure
  1. Veena Zamvar1,
  2. John W L Puntis1,
  3. Girish Gupte2,
  4. Gill Lazonby1,
  5. Christine Holden3,
  6. Elaine Sexton3,
  7. Christopher Bunford3,
  8. Susan Protheroe3,
  9. Susan V Beath2
  1. 1Department of Paediatric Gastroenterology, Leeds Children's Hospital, The General Infirmary at Leeds, Leeds, UK
  2. 2Department of Paediatric Hepatology and Small Bowel Transplantation, Birmingham Children's Hospital, Birmingham, UK
  3. 3Department of Paediatric Gastroenterology, Birmingham Children's Hospital, Birmingham, UK
  1. Correspondence to Dr Sue Beath, The Liver Unit (including Small Bowel Transplantation), Birmingham Children's Hospital, West Midlands B4 6NH UK;


Although most children discharged on home parenteral nutrition (HPN) will achieve enteral autonomy, some remain parenteral nutrition dependent; those who develop life-threatening complications may undergo small bowel transplantation (SBTx). The aim of this study was to investigate the relationship between social circumstances, compliance and complications.

Subjects and methods An observational study in 2008–2012 on 64 children (34 HPN, 30 SBTx) from three units (two regional gastroenterology; one transplant). Social circumstances were assessed routinely as part of discharge planning; adherence by families to home care management was scored, and episodes of catheter-related blood stream infection and graft rejection were recorded for 2 years and related to compliance and social circumstances.

Results A quarter of families had a disadvantaged parent: non-English speaking (n=11), unable to read (n=5), physical disability (n=3), mental health problems disclosed (n=10); 20% children were cared for by a lone parent. Discharge home was delayed by social factors (n=9) and need for rehousing (n=17, 27%). 17/34 (50%) of HPN and 12/30 (40%) of transplant families were assessed as fully adherent. 10 families were assessed as non-adherent, eight were subject to child protection review and care was taken over by another family member (n=3) or foster parents (n=2). The risk of catheter-related blood stream infection was increased by parental disadvantage and age <3 years (p<0.05). Poor compliance was associated with complications in HPN and SBTx recipients.

Conclusions Children receiving complex home care may be socially isolated and measures to support improved compliance such as increased community support, social care involvement and respite care may improve outcomes.

  • Gastroenterology
  • Multidisciplinary team-care
  • Comm Child Health
  • compliance
  • transplantation

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic

  • Home parenteral nutrition and small bowel transplantation are life-saving interventions.

  • Home care improves quality of life for the patient, but parents and families of chronically ill children experience high levels of stress.

  • Non-compliance with treatment and care routines is multifactorial, is more likely with increasing complexity of care needs and can have disastrous consequences.

What this study adds

  • Children receiving long-term home parenteral nutrition (HPN) and those progressing to small bowel transplantation (SBTx) are more likely to suffer serious complications when parents are disadvantaged or unsupported.

  • A simple scoring system may help identify children at risk from non-compliance with home care protocols.

  • Children in families with adverse social circumstances and/or poor compliance may benefit from targeted interventions that increase levels of support offered following discharge home.


Parenteral nutrition saves the lives of many children with intestinal failure. Technological advances and patient/family choice has driven the development of home care (home parenteral nutrition; HPN) and allowed children to grow and lead an otherwise normal life.1 HPN, however, demands strict attention to aseptic technique, monitoring routines, medication schedules and enteral nutritional support regimens in close conjunction with the hospital nutrition support team. Life threatening complications such as progressive liver disease are an indication for small bowel transplantation (SBTx),2–6 but a high degree of technical competence is also required from parents of children who have had SBTx owing to complex drug administration requirements and demanding care routines aimed at minimising drug related complications and detecting early signs of rejection. Despite many parents and carers succeeding in providing excellent long-term care,7–9 some children develop serious complications. The aim of this study was to explore whether there were identifiable social factors connected to poor compliance with management and follow-up protocols after discharge, and if these were associated with an increased risk of central venous catheter infection or graft rejection, by focusing on children with long-term parenteral nutrition (PN) dependency, and those undergoing SBTx.

Subjects and methods

This study was based on a retrospective review of case notes. Social circumstances were explored in detail by the supervising multidisciplinary team (MDT) as a routine part of the hospital discharge planning process. Patient demographics (age, gender, underlying diagnosis) are shown in table 1. Patients were recruited from two paediatric gastroenterology units (the General Infirmary at Leeds, n=14, and Birmingham Children's Hospital, n=20) and comprised all children with intestinal failure who fulfilled the following criteria: (A) established on HPN between 1998 and 2008; (B) 2 years of follow-up data available. In addition, all children undergoing SBTx in a supraregional centre (Birmingham Children's Hospital, n=30) who fulfilled the following criteria: (A) transplanted between 1998 and 2008; (B) 2 years of follow-up data available. Data on complications were obtained over the 2 year period following initial discharge from hospital. Children who discontinued HPN or died within 2 years of discharge were excluded.

Table 1

Demographic details of patients

The supervising MDT consisted of nurse specialists, and in some cases a social worker. Additionally the SBTx patients had a family support worker. Information included: single parent household, availability of extended family, parental disability or learning difficulties, psychiatric disorder, adequacy of housing, non-medical reasons for delay in discharge, parental separation during follow-up and child protection proceedings. A global assessment of adherence behaviour (‘compliance’) was agreed by at least three MDT members and was based on the following criteria: attendance of follow-up appointments, timely collection of prescriptions from pharmacy (within 48 h), attendance for monitoring blood tests, ease of telephone contact. Families were judged ‘fully compliant’; ‘partially compliant’ (missed >2 clinic appointments/year, required ≥2 reminders for blood tests, on >50% of occasions took >7 days to respond to urgent telephone messages) and ‘non-compliant’ (unable to comply with follow-up regimens; referral to a child protection team made—table 2). Adherence to management and follow-up protocols as above was given an arbitrary score that was then used as a proxy measure for compliance with MDT management recommendations.

Table 2

Compliance Scoring Grid (used to grade compliance)

Complications recorded were: (1) Catheter-related blood stream infection (CRBSI), defined as a positive blood culture in association with at least one of the following: fever >38.5 C, rigours or hypotension,10 no alternative identifiable source of infection; (2) Histologically confirmed rejection of small bowel allograft.


The relationship between complications, social circumstances and compliance was tested in a 2×2 contingency table using Fisher's exact test and two-tailed p values using software published by GraphPad Instat (Instat 3 for Macintosh July 2009).


Ethical approval was not considered necessary as this study was conducted as part of normal clinical audit processes in each unit and pooled data were anonymised.


The patients were all drawn from a similar cohort who had been exposed to the hazards of long-term PN; SBTx patients were older having received PN at an earlier point in their lives (SBTx is not carried out unless complications of HPN have occurred) see table 1. Gastroschisis as the underlying cause of intestinal failure was relatively more common in SBTx recipients (36%) compared with HPN patients (8.8%).

The social circumstances at the time of the survey in May 2009 are shown in table 3. The prevalence of single parent household and involvement in care giving by extended family was similar for HPN and SBTx patients. Disadvantage in one parent was recorded for a quarter of families and included: non-English speaking n=11; limited education (unable to read) n=5; physical limitation n=3. Mental health problems in a parent had been disclosed to a member of the MDT in 10 families. There were no differences between HPN and SBTx groups for the prevalence of disadvantage and mental health problems, but rehousing was required in a greater proportion of HPN children (41%). Discharge home was delayed by non-medical factors in five patients after training for HPN, and in four SBTx patients, because of concerns about the adequacy of housing (n=8) and difficulty registering with a general practitioner (n=1). The support available to families depended on the local arrangements: one HPN centre employed a home care company (a package with nursing support) and another used key workers in the community. The SBTx families were resident in all parts of the UK including Scotland and Northern Ireland and had variable levels of support with only 8/30 having input from a community paediatric nurse. Respite care was not generally available and was not accessed by any of the children on HPN or after SBTx during the course of this survey. Parental separation occurred in 3 (9%) of HPN families and in 5 (17%) of SBTx families during the 2 years of follow-up.

Table 3

Social circumstances of patients at time of first home discharge on HPN or after SBTx

Using the criteria given in table 2, 17/34 (50%) of HPN and 13/30 (40%) of SBTx recipients’ families were fully compliant with management and follow-up protocols, while 36–38% were partially compliant. For 4/34 (12%) HPN patients and for 6/30 (20%) SBTx patients, non-compliance was a major concern to the MDT. Ten children were alerted to social services, of whom eight children were subject to child protection review, resulting in the care of three being taken over by another family member and two going into foster care.

CRBSI occurred in 14/34 HPN patients (27 episodes) varying from 1 to 8 episodes in 1 year (1 every 269 days). In SBTx patients, acute rejection developed in 8/30 (table 4). The relationship between CRBSI or allograft rejection and social factors was assessed according to: single parent household; lack of support from extended family; presence of parental disability or disadvantage; mental distress where disclosed; need for rehousing; parental separation after discharge home; adherence with follow-up (table 4). Only parental disadvantage was related to sepsis risk for HPN patients (p<0.05). The diagnosis of enteropathy was associated with parental disadvantage in 6/9 patients (2/25 children with short bowel syndrome or dysmotility had a parent who was disadvantaged), but there was no association between CRSBI and enteropathy per se. In the HPN group, age less than 3 years was significantly associated with increased risk of CRBSI (p<0.04), but no effect of diagnosis, gender or gastroenterology centre on risk of CRBSI was identified. Mortality and rate of weaning from PN was also independent of gastroenterology centre. Other outcomes in the HPN group were as follows: 11 children were weaned from PN, 5 children on HPN died, 10 remained on PN and 8 progressed to small bowel transplant of whom 5 are still alive (as of 31 December 2012).

Table 4

Associations between social circumstances and complications occurring at home over 2 years (CRBSI in patients on HPN; episodes of histologically proven acute rejection in SBTx recipients within 2 years of transplantation)

For SBTx recipients, no social or demographic factors were strongly related, but there was a trend towards more acute rejection occurring where there was a lack of extended family support (p=0.07). In the SBTx group none developed chronic rejection during the 2 year observation period of this survey. The role of compliance in the development of complications was not found to be very strong in either group of patients, although there was a trend for poor compliance in HPN families to be related to CRBSI (p=0.08), and in SBTx recipients allograft rejection was slightly more increased in those families found to be poorly compliant compared with those who were fully compliant (p=0.09). When the HPN and SBTx groups are combined in the contingency tables, poor compliance became a more significant risk factor for medical complications (ie, CRBSI or acute rejection) with a p value<0.01 suggesting that sample size is a factor in this survey.


The social circumstances of the children in this survey were highly variable, some coming from two parent households with support from an extended family network and others living in isolated and deprived conditions. The incidence of single parent households for HPN patients and intestinal transplant patients (20%) was similar to that previously reported for HPN patients (25%)8; and to the UK national average (27%).11 In contrast, liver transplant recipients in the US state of Delaware reported 50% of single parent households,12 but there was a high level of support from the extended family (62.5%). This is likely to be important because there was a trend towards episodes of rejection in SBTx recipients occurring when family support was lacking (p=0.07). In our study, parental disadvantage was an important factor increasing risk of CRBSI (p<0.05). Parental disadvantage was found especially frequently in HPN patients with enteropathy (six of nine families had a disadvantaged parent with non-fluency in English (five), and one had cerebral palsy), but the role of ethnicity in disadvantage is unclear as five families were of Indian or Pakistani origin and four were Caucasian, and enteropathy in isolation was not associated with CRBSI. Mental ill health in a parent was mainly represented by depression and probably under-reported in this sample (16%) as we did not specifically screen for this problem.

We did not record the employment status of families, however, a large number needed to be rehoused (41% in the HPN group) prior to discharge, because their existing accommodation was inadequate. Parental separation rates over 5 years in the UK are reported to be 16%11 which is similar to that seen in our cohort (12.5% followed up over 2 years). This is surprisingly low given the stress reported by many parents caring for a chronically ill child,13–19 and suggests that our follow-up period may be too short, or this group has developed some useful coping strategies.

The overall rate of CRBSI is higher than international reports, possibly because most of the children being followed up were still young (24/34 aged <5 years), a known risk factor for sepsis; age <3 years was found to be a separate risk factor for CRBSI in this survey. In SBTx, early acute rejection affects 50–80% of recipients in the first 3 months after transplant, but the incidence reduces thereafter.20 The SBTx patients in this survey had a variable duration of follow-up, but we examined data in the first 2 years postdischarge when 26% of patients had developed rejection; similar to reported rates of rejection ranging from 5% to 23%21 depending on combination of organs transplanted and immune suppression. The underlying risk factors for allograft rejection are still being elucidated and include sensitisation of the recipient, human leukocyte antigen (HLA) mismatching, intercurrent viral infection, as well as non-compliance.

Although complications such as CRBSI and rejection of transplanted organs are associated with non-compliance, this subject is poorly defined and under-recognised by clinicians.12 ,2224 Parents’ sense of being blamed for complications, and underlying depression may also be an important factor15 ,25 ,26 and would need to be considered in any prospective interventional study. A study evaluating post-traumatic stress disorder in 19 liver transplant recipients22 found a correlation between high stress in parents and poor adherence indicated by erratic administration of antirejection drugs and in another study of children after liver transplant, parental non-compliance contributed to the majority of problems with liver dysfunction and graft loss.12 We chose to use a simple measure based on familiarity between MDT and the parent; we did not attempt to record adherence with intravenous fluid or drug administration, although these empirical measures would have been informative

At the time of this study we found that despite the complexity of care and requirements for regular and frequent monitoring, 50% of HPN and 44% of small bowel transplant patients and families were judged fully compliant. Lack of co-operation with treatment protocol, either through disadvantage/disability, little family support or poor compliance, resulted in more complications in both groups. In eight children, child protection procedures became necessary and resulted in increased supervision by the medical team and social workers and was associated with improved compliance for all the families and children who were considered non-compliant. This suggests that lack of support for families combined with parents’ ignorance or denial about the impact of non-compliance was an important factor, and supports the notion that interventions based on the information-motivation-behavioral skills model would improve compliance in future.24 ,27 ,28 This observation needs testing in a prospective multimethod study incorporating cognitive factors, interpersonal factors between patient, family and team members, and patient factors such as depression and cultural attitudes.26

The cohort of children surveyed represents the most challenging group within the Birmingham and Leeds home PN programmes. By using the inclusion criterion of at least 2 years’ treatment with HPN, children with relatively poor prognosis were selected for study: only a third could be weaned from PN and in the long term 13 (38%) either died or required a SBTx. This compares with the overall rate of weaning from PN in children with short bowel syndrome of 75–90% in healthcare systems with established nutritional care teams.29 Given the unrelenting demands of providing complex nursing care while also carrying out normal parental roles, it is not surprising that some families are overwhelmed and exhausted.30 The care of children with complex conditions who require comprehensive nursing care often involving medical technology has increasingly shifted from hospital to home in the past 20 years. Potential benefits include provision of a more normal environment and better quality of life for the child, and a reduction in costs to the health service. However, this cost saving often comes at the expense of the parents and carers who frequently need to give up jobs, free time, hobbies and holidays.13 ,31 While socially disadvantaged families can readily be identified, it is disturbing to note that support needs may have been underestimated from the outset. The reason for this could relate to reluctance on the part of families to admit they need help, compounded by resource restrictions applied by the local health provider.

In this study the families generally coped extraordinarily well with the demands placed on them, but the true cost to parents and their children is often unrecognised and in some cases this leads to disastrous treatment failure because parents have simply been overburdened.30 Lack of compliance is hardly ever the only factor, and compliance is not simply a ‘patient problem’; health providers also have a major role to play in the way they improve access to advice and support. It is illogical to invest in HPN and SBTx if things then go wrong at home because of a failure to secure appropriate levels of support.


There is an urgent need to identify potential stresses and support healthy coping strategies in high-risk families, although this retrospective study cannot answer the question of which interventions would be most helpful to families in the UK setting. However, we have developed a potential tool (the compliance scoring grid—table 2) to monitor patients and families who might be at greater risk of medical complications because of their adverse social circumstances, that could be used in a multimethod approach26 to target those whose children are at increased risk of serious adverse outcomes. The results of our survey suggest that a sufficiently powered study would need at least 64 subjects to detect statistically significant effects. Those families who are only partially compliant with follow-up are most likely to benefit from increased support from outreach or community workers and a Common Assessment Framework approach to establish which area of care would be of practical use to them.17 ,24 ,28 We noted that CRBSI/graft rejection tended to occur more often when a parent was disadvantaged by having disability, and, when both groups (HPN and SBTx) were combined for the purposes of statistical analysis, poor compliance with clinical management advice and follow-up protocols was significantly associated with CRBSI/graft rejection. We conclude that a prospective study evaluating how such factors as early social care involvement, access to psychology advice,22 increased community nurse input, home nursing care, respite care, hospice access, flexible outpatient appointments12 and specialist outreach clinics24 ,28 could influence outcomes, is justified.


The authors are grateful for the advice and invaluable support provided by all the multiprofessional team members of the Small Bowel Transplant programme of Birmingham Children's Hospital especially Cherry Clifford, Clare James, Debbie Hart, Lindsay Hogg, Julie Taylor and Velma Wright. The authors also wish to thank Dr MS Murphy for his insights and reflections on the experiences of children with intestinal failure and their families at the outset of this project.



  • Third Revision inc Editors changes—5 December 2013

  • Contributors VZ collected some of the data, presented the results at scientific meetings and wrote the initial drafts and collaborated with the writing of subsequent drafts. GG made a substantial contribution to the design of the study, analysis and interpretation of the data and contributed to the writing of the manuscript especially the discussion. JWLP contributed substantially to the analysis and interpretation of data, critically revised the manuscript several times and approved the final version. GL, CH, ES and CB collected some of the data, provided compliance scores and collaborated with the writing of manuscript. SP made a substantial contribution to the design of the study, analysis and interpretation of the data especially the Home PN patients and commented on the manuscript. SVB contributed to the design of the study, populated the database, performed statistical analysis and revised later drafts of the manuscript and approved the final version.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.