For a number of reasons it is important to identify and treat childhood tuberculosis (TB) in order to control the disease. Firstly, it acts as a marker of recent disease transmission and secondly, it is a reservoir of infection for the future. However, children’s services may have been neglected because most TB occurs in adults (in the UK only 5% of reported cases occur in children aged 0–14 years)1 and therefore the impetus for improvement in service provision may have been driven by adult physicians who have focused on disease in adults.2 The low risk of contracting the disease from children means they have been seen as low priority when it comes to developing services. Moreover, other key issues are under-estimation of the disease burden, due to difficulties in confirming the diagnosis with standard microbiological techniques, and under-notification.3

Guidance for managing TB has recently been published by NICE.4 This suggests that TB in children should be managed either by a paediatrician with experience and training in the treatment of TB, or by a general paediatrician with advice from a specialised physician. Our experience locally was that this model of care was often not followed.

The aim of this study was to determine the various models of care and service provision for children with TB living in the UK.

METHODS

A postal questionnaire was sent to the most appropriate paediatrician and adult physician in all acute trusts seeing children in the UK. A pilot survey was initially carried out in London5 and then replicated throughout the rest of the UK. If there was no reply after an appropriate delay, a further questionnaire was sent. We asked about inpatient and outpatient services for children with TB, as well as about services for children in contact with TB.

The most appropriate paediatrician was identified by contacting clinical leads in all trusts with paediatric departments, from a list provided by the Royal College of Paediatrics and Child Health.

The respiratory physician was identified from a list provided by the Royal College of Physicians, as well as by sending duplicate questionnaires to the paediatricians contacted, to be forwarded to the most appropriate adult physician involved in the care of these children in their trust.

We attempted to outline service provision in three main areas: (1) inpatient care, (2) outpatient care and (3) contact management.

RESULTS

There was a 96% response rate, with a total of 323 responses received from 199 of the 205 trusts contacted. Of these, 184 (57%) were from paediatricians and 139 (43%) from adult respiratory physicians.

The number of children with TB reported to be seen in each trust per year varied widely (median 1.5 per year, range 0–30).

Inpatient management

Children with TB were admitted to paediatric wards in 99% of trusts, while 1% said they would be admitted to adult wards. Paediatricians and physicians were in strong agreement as to what type of ward children were admitted to. All paediatricians and 97% of physicians said that these children were likely to be admitted to paediatric wards (table 1). Adult physicians felt that children would be transferred to adult wards if isolation facilities were required.

Children admitted with TB were managed by paediatricians in 141 (71%) trusts, physicians in 11 (5%) trusts or jointly in 47 (24%) trusts. Where there was joint management, this was a formalised arrangement in only 25 (53%) trusts. In most trusts children were managed by general (127 (64%) trusts) or respiratory paediatricians (51 (26%) trusts).

Overall, 132 trusts (66%) had a named consultant with overall responsibility for childhood TB. The responses from 106 (58%) paediatricians indicated there was a named consultant for children with TB compared with 75 (54%) adult physician responses, although in only 50 (25%) trusts were there consistent responses from both paediatricians and physicians.

Negative pressure rooms were reported to be available for children in only 42 trusts (21%). In 21 (11%) of these, children would have had to be transferred to adult wards as the facilities were not available on the paediatric wards.

Outpatient management

Children were seen in paediatric clinics in 163 (82%) trusts, while in a further 28 (14%) they were seen in both paediatric and adult clinics. Only 10 (5%) trusts had instituted designated family clinics for TB.

Paediatricians and physicians agreed that the majority of children were seen in paediatric clinics, with just over 10% stating they were seen in both paediatric and adult clinics. In addition, 20% of physicians reported that children with TB were seen in adult clinics only, while a much smaller proportion of paediatricians (9%) reported this to be the case (table 1).

Overall, joint outpatient care by a paediatrician and a physician was reported by 205 (63%) responses but less than a third of these stated this was on a formal basis. Joint outpatient care was offered by 94 (47%) trusts, but this was a formalised arrangement in only 26 (13%).

A directly observed therapy service (DOTS) for children was provided by 116 (58%) trusts, with the proportion being highest in the greater London area (73% compared to 54% of trusts elsewhere).

A TB nurse was available in 161 (81%) trusts in the UK, although once again trusts within London did better with 24 (92%) having nurses involved in the management of childhood TB compared with 138 (80%) in the rest of the UK.

Contact management

Both paediatricians and adult physicians felt they saw the majority of childhood contacts (table 2 provides details of responses from outside London).

Children in contact with TB in the UK were managed by paediatricians in 81 trusts (41%) and by physicians in 67 trusts (33%). In a further 51 (26%) trusts, contacts were managed by both, while respondents from three trusts were not aware of arrangements for contact management.

DISCUSSION

This survey demonstrated widely varying models of care for children with TB in the UK, probably reflecting demographic and geographical variations in this infection.

Many areas see few children with TB, especially outside London. While 35% of childhood TB occurs in London, the incidence in other parts of the UK is much lower (8/100 000 vs 2.4/100 000).6 Many areas of the UK are therefore unlikely to gain adequate experience to optimally manage childhood TB.

Most trusts admitted children with TB to general paediatric wards, where they are looked after by general paediatricians, and most outpatients were seen in paediatric clinics. Despite this, only 66% of trusts had a named consultant for children with TB; moreover, both paediatricians and physicians were aware of this arrangement in only a quarter of trusts.

Few trusts have negative pressure isolation facilities that can be used for children. One argument for not making this service available is that most cases children with TB are not infectious. However, 15% of parents/visitors to a child admitted to hospital with TB could have undiagnosed TB.4 It may thus be necessary to isolate children with TB if parents/visitors are at risk of multi-drug resistant TB. Such rooms should also be available for older children at risk of multi-drug resistant TB who are able to transmit the disease to others.

At the time of our survey, guidance for the clinical care of these children came from the Joint Tuberculosis Committee of the British Thoracic Society. This recommended that children with TB should be managed by a paediatrician in collaboration with a clinician with TB experience (often an adult respiratory physician).7 8 This has been restated in the recent NICE guidelines.4

Our survey suggests that this guidance may not be followed, since in over 70% of trusts children with TB were looked after by paediatricians alone. There are few family TB clinics, but this is perhaps not surprising given the low numbers of children with TB seen across all trusts. However, arrangements for children with TB to be managed jointly by those with experience in paediatrics and those experienced in TB should be available to all children throughout the UK.

The implications of our study are that since most trusts see few children with TB, a clinical service network for TB should be developed so that expert advice on childhood TB is available wherever the child is seen. Networks can also provide appropriate isolation facilities for children when required.

A recent government action plan has also recommended improving the consistency of clinical care throughout the UK.9 This includes setting up clinical service networks according to recommendations from the Children’s NSF, with the institution of designated local TB co-ordinators and the creation of “clinical care pathways” with named case managers assigned to every TB patient. It is proposed that there should be clear identification and facilitation of access to appropriate isolation facilities.

The establishment of family clinics and joint services for patients co-infected with HIV and routinely supervised continuing care and increased use of DOTS should also be implemented.

More recently a toolkit for planning and developing TB services has been published by the Department of Health.10 It sets out a framework for assessing local needs, to enable primary care trust commissioners plan and develop satisfactory TB services for local populations and subsequently monitor their delivery.

These recommendations may improve the care provided to children with TB both within and outside London.