Abstract

Objective. To identify the ideal programme of transitional care for adolescents with juvenile idiopathic arthritis (JIA) as perceived by users and providers, and to examine the feasibility of achieving this within a UK National Health Service context.

Methods. A modified two-stage Delphi study was undertaken with rheumatology health professionals, young people with JIA (aged 12–25 yr) and their parents. Participants were presented with statements about transitional care and asked to rate (i) the extent to which these constituted best practice, and (ii) their feasibility.

Results. Second-round questionnaires were completed by 83 individuals, representing an overall response rate of 90%. Items strongly agreed to constitute best practice and highly feasible included: ‘addressing young people's psychosocial and educational/vocational needs’; ‘using an individualized approach’; ‘providing honest explanations of the adolescent's condition and health-care’; ‘providing opportunities for adolescents to express opinions and make informed decisions’; ‘having continuity in health personnel’; and ‘giving adolescents the option of being seen by professionals without their parents’. However, providing adolescent-focused environments, professionals knowledgeable in transitional care and opportunities for young people with JIA to meet similar others were seen as feasible in only a few hospitals.

Conclusions. There is considerable agreement as to the most important elements of transitional care. Those that are easily achievable should be undertaken in all hospitals that care for adolescents with JIA. However, not all elements identified were perceived as easily feasible; further research is required to determine how to implement these elements.

The impetus towards transitional care for young people with chronic illness in the UK has been strengthened recently by the National Service Framework (NSF) for Children [1] and an intercollegiate report [2]. Transitional care is also supported by the American Academy of Pediatrics, who maintain that the goals of transitional care are to ‘maximize lifelong functioning and potential through provision of high-quality, developmentally appropriate health-care services that continue uninterrupted as the individual moves from adolescence to adulthood’ [3].

The need for transitional care in rheumatology is compelling. Over a quarter of juvenile idiopathic arthritis (JIA) cases begin during adolescence and the impact of chronic illness on adolescent development has been well documented [4]. Significant morbidities have also been reported during adulthood [4–8]. Unfortunately, there is little evidence of what constitutes the optimum service for transition in any chronic childhood-onset illness [9, 10]. In the absence of empirical evidence, the development of such services must draw upon the expertise of ‘legitimate stakeholders’, including professionals and patients [11, 12]. This not only ensures that the interests of such people are represented, but also increases ownership of the findings and their influence on practice and policy [11]. Recognizing this, the present study sought to use the Delphi method to determine expert opinions regarding best practice in transitional care and the feasibility of providing such services within a UK context.

Methods

The Delphi method typically involves a series of data-gathering rounds in which panel members independently offer and refine their opinions regarding a number of defined statements [12, 13]. Researchers strive to determine the extent to which respondents agree with the statements under consideration and to assess the extent to which respondents agree with one another [12, 13].

To maximize validity, the Delphi method requires that the panel consist of experts [12]. There is no set criterion as to what defines ‘expert’ status. However, when chosen carefully, the views of a small number of people can be representative of a target group [11, 14]. It is thought that whilst reliability of the results may decline rapidly with fewer than six panel members, improvements in reliability are relatively small in groups larger than 15 [12].

The panel

The panel consisted of ‘providers’ [rheumatology health professionals, including rheumatologists and allied health professionals (AHP)] and ‘users’ (young people with JIA and their parents).

Providers were selected using a two-stage process: (i) the British Paediatric Rheumatology Group (BPRG) were asked to identify individuals from their membership list who were currently practising in rheumatology. To allow for attrition between rounds, researchers then purposefully sampled 23 consultants and 23 AHPs to include professionals from a range of adult and paediatric settings (including specialist and non-specialist centres) and to reflect geographical diversity.

The panel of users was derived from the membership list of the Children's Chronic Arthritis Association (CCAA), who, in order to protect anonymity, were asked to recruit approximately 20 eligible young people with JIA diagnosed for a least 1 yr, aged between 12 and 25 yr, and a parent.

Participants were informed about the nature of the study and told that participation would require them to complete a questionnaire on at least two separate occasions and within specified deadlines. Three individuals (AHPs) felt unable to continue at this stage.

The study had approval from the West Midlands Multicentre Research ethics committee.

Questionnaire

To reduce the numbers of rounds in our study, the questionnaire items were generated from the findings of an earlier needs assessment of transitional needs [15, 16].

Data were collected using a 26-item questionnaire designed for self-completion (Table 2). This was pretested (n = 4) to ensure comprehension and user-friendliness. To increase face validity, young people and their parents were also given a glossary of terms. All panel members were asked to indicate the extent to which they agreed that each statement constituted best care. Health professionals were also asked to indicate the feasibility of each statement. Five-point response scales were used (Table 2) and members were encouraged to provide written clarification of their responses. To maximize independence, young people and their parents were asked to complete their questionnaires separately.

Additional data were collected to verify the Panel's expert status. Providers were asked to indicate how long they had worked in rheumatology and how many adolescents they saw per month. Young people were asked to indicate their disease duration and how long they had been in the care of rheumatology services.

Procedure

The questionnaire was distributed to all panel members (Round 1). On return, the responses of each subgroup were collated and sent back with a graphical indication of the group's median scores, interquartile ranges and written clarifications. Panel members were subsequently asked to re-score the items in the light of the group's initial responses (Round 2). Reminders were sent to non-respondents.

Analyses

Statistical analyses were performed using SPSS 10.0 (SPSS, Chicago, IL, USA). Analyses were undertaken to determine whether the sample constituted one homogeneous panel or several subpanels. Group differences were analysed using Kruskal–Wallis and Mann–Whitney tests where appropriate. The level of support for each item was indicated by the median and the level of consensus within each subpanel shown by the mean absolute deviation from the median. The impact of the Delphi technique on final agreement and consensus was examined by assessing change between the first and second rounds using the Wilcoxon's signed ranks test. Statistical significance was set at the 0.01 level to adjust for multiple testing.

Results

Participant characteristics

The final panel consisted of 74 members and represents a final response rate of 90% (Table 1).

Table 1.

Panel members

 Round 1 n (%) Round 2 n (%) Response rate (%) 
Providers    
    Adult rheumatologists 11 (25) 11 (26) 100 
    Paediatric rheumatologists 10 (23) 10 (24) 100 
    Paediatricians 2 (5) 2 (5) 100 
    Physiotherapists 8 (19) 8 (19) 100 
    Occupational therapists 7 (16) 6 (14) 86 
    Nurses 5 (12) 5 (12) 100 
    Total providers 43 (100) 42 (100) 98 
Users    
    Young people with JIA 20 (51) 16 (50) 80 
    Parents 19 (49) 16 (50) 84 
    Total users 39 (100) 32 (100) 82 
Total panel 82 (100) 74 (100) 90 
 Round 1 n (%) Round 2 n (%) Response rate (%) 
Providers    
    Adult rheumatologists 11 (25) 11 (26) 100 
    Paediatric rheumatologists 10 (23) 10 (24) 100 
    Paediatricians 2 (5) 2 (5) 100 
    Physiotherapists 8 (19) 8 (19) 100 
    Occupational therapists 7 (16) 6 (14) 86 
    Nurses 5 (12) 5 (12) 100 
    Total providers 43 (100) 42 (100) 98 
Users    
    Young people with JIA 20 (51) 16 (50) 80 
    Parents 19 (49) 16 (50) 84 
    Total users 39 (100) 32 (100) 82 
Total panel 82 (100) 74 (100) 90 

As a group, the providers had worked in rheumatology for a median of 10 (range 1.5–31.0) yr and currently saw 10.0 (range <1.0–70.0) adolescent patients per month. Young people had a median age of 15.4 (range 12.0–21.1) yr, a self-reported disease duration of 13.0 (range 3.0–20.0) yr and had been in the care of a specialist (i.e. rheumatologist, paediatrician) for their arthritis for 5.5 (range 2–17) yr. Most parents (85%) were mothers.

One panel or several?

Analyses revealed that the participants did not constitute one homogeneous panel, but several subpanels. Only the parent and young people's groups appeared similar in their mean rankings (with the exception that young people agreed more strongly than parents that ‘adolescents should be allowed to decide who is invited into the consultation room’; P = 0.003). It was decided, therefore, that these two groups should be combined to make one composite user panel. However, as their familial relationship means that the two groups could not be considered truly independent, the mean score of each familial pairing was used as a composite user score. Consequently, from this point forward, the Delphi panel comprised three subpanels: consultant, AHP and user.

Best practice

Each group agreed that most items constituted best practice and generally had a high degree of consensus within each group (Table 2). Between-group analyses, however, demonstrated a number of significant differences in the extent to which items were rated as best practice (Table 3). In general, AHPs rated items higher than both the consultants and the users, and users frequently had the lowest median scores. However, for the most part, these differences were only in the strength of agreement (i.e. ‘strongly agree’ as opposed to ‘agree’).

Table 2.

Level of agreement and consensus regarding the extent to which items constitute best practice and are feasible

 Best practice
 
  Feasibility
 
 
Item Consultants AHPs Users Consultants AHPs 
Health-care teams need to provide honest explanations of the adolescent's condition and associated health-care 1.0 (0.04) 1.0 (0.05) 1.0 (0.12) 2.0 (0.65) 2.0 (0.26) 
Health professionals have a responsibility to always provide opportunities for adolescents to express opinions and make informed decisions 1.0 (0.30) 1.0 (0.37) 1.0 (0.15) 1.0 (0.15) 1.0 (0.21) 
Addressing young people's psychosocial and educational/vocational needs is as important as addressing their medical needs 1.0 (0.30) 1.0 (0.05) 1.5 (0.44) 2.0 (0.43) 2.0 (0.32) 
Continuity in health personnel is essential in building the adolescents’ trust, confidence and willingness to disclose 1.0 (0.52) 1.0 (0.11) 1.5 (0.32) 2.0 (0.25) 2.0 (0.32) 
Adolescents should always be given the option of being seen by professionals without their parents 1.0 (0.13) 1.0 (0.47) 1.5 (0.38) 2.0 (0.52) 1.0 (0.32) 
Transitional care should be delivered using an individualized approach 1.0 (0.48) 1.0 (0.21) 1.5 (0.32) 2.0 (0.24) 1.0 (0.53) 
Transitional care should be developmentally appropriate 1.0 (0.27) 1.0 (0.42) 2.0 (0.15) 2.0 (0.10) 2.0 (0.32) 
Adolescents and parents should be given free and open access to all available information regarding JIA and related health issues 2.0 (0.30) 1.0 (0.26) 1.0 (0.29) 2.0 (0.43) 2.0 (0.44) 
Facilitating adolescents’ self-advocacy skills (e.g. decision-making, effective communication, disclosure) is an important aspect of transitional care 1.0 (0.45) 1.0 (0.16) 2.0 (0.29) 2.0 (0.30) 2.0 (0.79) 
Adolescents should be allowed to decide who is invited into the consultation room 2.0 (0.48) 1.0 (0.26) 1.5 (0.32) 2.0 (0.19) 2.0 (0.42) 
The transitional needs of adolescents with JIA should be reviewed on at least a yearly basis 2.0 (0.26) 1.0 (0.26) 2.0 (0.32) 2.0 (0.24) 1.0 (0.42) 
Optimal transfer depends upon close co-operation between paediatric services and adult services 2.0 (0.26) 1.0 (0.16) 2.0 (0.35) 2.0 (0.52) 2.0 (0.47) 
Strong interagency links (e.g. health, education, social services, community services) are an essential element of transitional care 2.0 (0.52) 1.0 (0.16) 2.0 (0.50) 2.0 (0.25) 2.0 (0.32) 
Optimal transitional care requires that health teams provide general and specific information to other professionals involved with adolescents with JIA 1.0 (0.48) 2.0 (0.47) 2.0 (0.24) 2.0 (0.55) 2.0 (0.37) 
Transitional care should be coordinated by a named member of the health-care team 2.0 (0.48) 1.0 (0.53) 2.0 (0.18) 2.0 (0.25) 2.0 (0.79) 
Transitional care is best provided by professionals who are knowledgeable about adolescent development 2.0 (0.35) 1.0 (0.47) 2.0 (0.62) 3.0 (0.75) 3.0 (0.63) 
Optimal transitional care involves a planned, gradual, withdrawal of parents from the primary position in the consultation 2.0 (0.22) 1.0 (0.26) 2.5 (0.50) 1.0 (0.29) 1.0 (0.21) 
Multidisciplinary teams are essential to the best care of adolescents with JIA 2.0 (0.57) 1.0 (0.11) 2.5 (0.41) 4.0 (0.55) 2.0 (0.63) 
Addressing parental needs is essential to supporting adolescent's movement towards normal adult functioning 2.0 (0.13) 2.0 (0.32) 2.0 (0.38) 2.0 (0.50) 2.0 (0.11) 
Proving adolescents with opportunities to meet similar others is in the remit of the rheumatology team 2.0 (0.17) 2.0 (0.37) 2.0 (0.50) 4.0 (0.19) 4.0 (0.33) 
Transitional care is best provided in dedicated adolescent environments (e.g. adolescent waiting areas) 2.0 (0.35) 2.0 (0.42) 2.0 (0.32) 4.0 (0.52) 4.0 (0.61) 
The optimal physical environment for providing transitional care is one that is age-appropriate 2.0 (0.35) 2.0 (0.53) 2.0 (0.26) 3.0 (0.52) 3.0 (0.63) 
Generic health issues should be addressed by the rheumatology health team 2.0 (0.39) 2.0 (0.47) 2.5 (0.39) 1.0 (0.40) 1.0 (0.21) 
Adolescents and parents need early preparation for their transfer to adult rheumatology services (e.g. starting at 11 yr) 2.0 (0.43) 2.0 (0.63) 2.5 (0.65) 3.0 (0.52) 2.0 (0.37) 
Transitional care needs to be based on anticipatory planning 2.0 (0.26) 2.0 (0.47) 3.0 (0.35) 2.0 (0.24) 2.0 (0.53) 
If adolescents are seen alone, parents must always be offered concurrent visits 3.0 (0.43) 2.0 (0.44) 2.0 (0.53) 2.0 (0.43) 1.0 (0.32) 
 Best practice
 
  Feasibility
 
 
Item Consultants AHPs Users Consultants AHPs 
Health-care teams need to provide honest explanations of the adolescent's condition and associated health-care 1.0 (0.04) 1.0 (0.05) 1.0 (0.12) 2.0 (0.65) 2.0 (0.26) 
Health professionals have a responsibility to always provide opportunities for adolescents to express opinions and make informed decisions 1.0 (0.30) 1.0 (0.37) 1.0 (0.15) 1.0 (0.15) 1.0 (0.21) 
Addressing young people's psychosocial and educational/vocational needs is as important as addressing their medical needs 1.0 (0.30) 1.0 (0.05) 1.5 (0.44) 2.0 (0.43) 2.0 (0.32) 
Continuity in health personnel is essential in building the adolescents’ trust, confidence and willingness to disclose 1.0 (0.52) 1.0 (0.11) 1.5 (0.32) 2.0 (0.25) 2.0 (0.32) 
Adolescents should always be given the option of being seen by professionals without their parents 1.0 (0.13) 1.0 (0.47) 1.5 (0.38) 2.0 (0.52) 1.0 (0.32) 
Transitional care should be delivered using an individualized approach 1.0 (0.48) 1.0 (0.21) 1.5 (0.32) 2.0 (0.24) 1.0 (0.53) 
Transitional care should be developmentally appropriate 1.0 (0.27) 1.0 (0.42) 2.0 (0.15) 2.0 (0.10) 2.0 (0.32) 
Adolescents and parents should be given free and open access to all available information regarding JIA and related health issues 2.0 (0.30) 1.0 (0.26) 1.0 (0.29) 2.0 (0.43) 2.0 (0.44) 
Facilitating adolescents’ self-advocacy skills (e.g. decision-making, effective communication, disclosure) is an important aspect of transitional care 1.0 (0.45) 1.0 (0.16) 2.0 (0.29) 2.0 (0.30) 2.0 (0.79) 
Adolescents should be allowed to decide who is invited into the consultation room 2.0 (0.48) 1.0 (0.26) 1.5 (0.32) 2.0 (0.19) 2.0 (0.42) 
The transitional needs of adolescents with JIA should be reviewed on at least a yearly basis 2.0 (0.26) 1.0 (0.26) 2.0 (0.32) 2.0 (0.24) 1.0 (0.42) 
Optimal transfer depends upon close co-operation between paediatric services and adult services 2.0 (0.26) 1.0 (0.16) 2.0 (0.35) 2.0 (0.52) 2.0 (0.47) 
Strong interagency links (e.g. health, education, social services, community services) are an essential element of transitional care 2.0 (0.52) 1.0 (0.16) 2.0 (0.50) 2.0 (0.25) 2.0 (0.32) 
Optimal transitional care requires that health teams provide general and specific information to other professionals involved with adolescents with JIA 1.0 (0.48) 2.0 (0.47) 2.0 (0.24) 2.0 (0.55) 2.0 (0.37) 
Transitional care should be coordinated by a named member of the health-care team 2.0 (0.48) 1.0 (0.53) 2.0 (0.18) 2.0 (0.25) 2.0 (0.79) 
Transitional care is best provided by professionals who are knowledgeable about adolescent development 2.0 (0.35) 1.0 (0.47) 2.0 (0.62) 3.0 (0.75) 3.0 (0.63) 
Optimal transitional care involves a planned, gradual, withdrawal of parents from the primary position in the consultation 2.0 (0.22) 1.0 (0.26) 2.5 (0.50) 1.0 (0.29) 1.0 (0.21) 
Multidisciplinary teams are essential to the best care of adolescents with JIA 2.0 (0.57) 1.0 (0.11) 2.5 (0.41) 4.0 (0.55) 2.0 (0.63) 
Addressing parental needs is essential to supporting adolescent's movement towards normal adult functioning 2.0 (0.13) 2.0 (0.32) 2.0 (0.38) 2.0 (0.50) 2.0 (0.11) 
Proving adolescents with opportunities to meet similar others is in the remit of the rheumatology team 2.0 (0.17) 2.0 (0.37) 2.0 (0.50) 4.0 (0.19) 4.0 (0.33) 
Transitional care is best provided in dedicated adolescent environments (e.g. adolescent waiting areas) 2.0 (0.35) 2.0 (0.42) 2.0 (0.32) 4.0 (0.52) 4.0 (0.61) 
The optimal physical environment for providing transitional care is one that is age-appropriate 2.0 (0.35) 2.0 (0.53) 2.0 (0.26) 3.0 (0.52) 3.0 (0.63) 
Generic health issues should be addressed by the rheumatology health team 2.0 (0.39) 2.0 (0.47) 2.5 (0.39) 1.0 (0.40) 1.0 (0.21) 
Adolescents and parents need early preparation for their transfer to adult rheumatology services (e.g. starting at 11 yr) 2.0 (0.43) 2.0 (0.63) 2.5 (0.65) 3.0 (0.52) 2.0 (0.37) 
Transitional care needs to be based on anticipatory planning 2.0 (0.26) 2.0 (0.47) 3.0 (0.35) 2.0 (0.24) 2.0 (0.53) 
If adolescents are seen alone, parents must always be offered concurrent visits 3.0 (0.43) 2.0 (0.44) 2.0 (0.53) 2.0 (0.43) 1.0 (0.32) 

Scores are Round-2 medians (mean absolute deviation from the median) and are ranked according to overall best practice. Best practice: 1, strongly agree; 2, agree; 3, neutral; 4, disagree; 5, strongly disagree. Feasibility: 1, feasible in all hospitals; 2, feasible in most hospitals; 3, feasible in some hospitals; 4, feasible in few hospitals; 5, feasible in no hospitals.

Table 3.

Between-group differences for items of best practice

 P 
AHPs agreed more strongly than consultants on  
    Annual review of individual patients 0.002 
    Multidisciplinary teams 0.002 
    Named coordinator 0.001 
    Strong interagency links 0.001 
    Continuity in personnel for consecutive visits 0.010 
    Giving adolescents free and open access to information 0.005 
    Planned withdrawal of parents from consultation 0.001 
    Concurrent visits for parents if adolescent seen alone* 0.010 
    Close interprofessional cooperation at transfer 0.000 
Consultants and AHPs agreed more strongly than users on  
    Anticipatory planning* 0.000 
    Developmentally appropriate care 0.000, 0.004 respectively 
    Multidisciplinary teams 0.004, 0.000 respectively 
    Generic health issues 0.000 
    Planned withdrawal of parents from consultation 0.000 
Consultants agreed more strongly than users on  
    Adolescents being seen without parents 0.000 
    Information provision to other professionals 0.005 
AHPs agreed more strongly than users on  
    Addressing psychosocial and educational/ vocational needs 0.002 
    Annual review of individual patients 0.003 
    Delivered by knowledgeable professionals 0.002 
    Strong interagency links 0.000 
    Continuity in personnel 0.009 
    Facilitating self-advocacy for adolescents 0.000 
    Close interprofessional cooperation at transfer 0.003 
Parents agreed more strongly than consultants on  
    Concurrent visits for parents if adolescents seen alone 0.008 
 P 
AHPs agreed more strongly than consultants on  
    Annual review of individual patients 0.002 
    Multidisciplinary teams 0.002 
    Named coordinator 0.001 
    Strong interagency links 0.001 
    Continuity in personnel for consecutive visits 0.010 
    Giving adolescents free and open access to information 0.005 
    Planned withdrawal of parents from consultation 0.001 
    Concurrent visits for parents if adolescent seen alone* 0.010 
    Close interprofessional cooperation at transfer 0.000 
Consultants and AHPs agreed more strongly than users on  
    Anticipatory planning* 0.000 
    Developmentally appropriate care 0.000, 0.004 respectively 
    Multidisciplinary teams 0.004, 0.000 respectively 
    Generic health issues 0.000 
    Planned withdrawal of parents from consultation 0.000 
Consultants agreed more strongly than users on  
    Adolescents being seen without parents 0.000 
    Information provision to other professionals 0.005 
AHPs agreed more strongly than users on  
    Addressing psychosocial and educational/ vocational needs 0.002 
    Annual review of individual patients 0.003 
    Delivered by knowledgeable professionals 0.002 
    Strong interagency links 0.000 
    Continuity in personnel 0.009 
    Facilitating self-advocacy for adolescents 0.000 
    Close interprofessional cooperation at transfer 0.003 
Parents agreed more strongly than consultants on  
    Concurrent visits for parents if adolescents seen alone 0.008 

All tests used the Mann–Whitney U-test.

*Median group scores differed between ‘agree’ and ‘neutral’ as opposed to ‘agree’ and ‘strongly agree’.

Feasibility

Most groups rated the items to be feasible in most, if not all, hospitals and generally demonstrated high levels of consensus (Table 2). However, both AHPs and consultants felt that ‘dedicated adolescent environments’ and ‘providing opportunities to meet similar others’ were feasible in only a few hospitals. Professionals also reported limited feasibility regarding transitional care being provided by ‘professionals who are knowledgeable about adolescent development’ or providing ‘age-appropriate’ environments.

Between-group analyses showed few significant differences in the ratings of feasibility, although AHPs rated it more feasible to provide ‘transitional care that is delivered by a multidisciplinary team’ (P = 0.002) and ‘early preparation for transfer to adult rheumatology services’ (P = 0.001) than consultants.

Best practice versus feasibility

Across the groups, several components of care were strongly agreed to constitute best practice. However, only some were deemed feasible in all/most hospitals (Table 4). Others were perceived as having much less feasibility (Table 5).

Table 4.

Components of transitional care attainable in most centres

Aspects of care strongly agreed to be best practice and deemed highly feasible 
Address young people's psychosocial and educational/vocational needs 
Use an individualized approach 
Provide honest explanations of the adolescent's condition and associated health-care 
Provide opportunities for adolescents to express opinions and make informed decisions 
Have continuity in health personnel 
Give adolescents the option of being seen by professionals without their parents 
Aspects of care strongly agreed to be best practice and deemed highly feasible 
Address young people's psychosocial and educational/vocational needs 
Use an individualized approach 
Provide honest explanations of the adolescent's condition and associated health-care 
Provide opportunities for adolescents to express opinions and make informed decisions 
Have continuity in health personnel 
Give adolescents the option of being seen by professionals without their parents 
Table 5.

Components of transitional care requiring further attention

Aspects of care agreed to be best practice, but deemed to have limited feasibility 
Multidisciplinary teams (consultants only) 
Professionals who are knowledgeable about adolescent development 
Age-appropriate physical environment 
Dedicated adolescent environments (e.g. adolescent waiting areas) 
Providing opportunities to meet similar others 
Aspects of care agreed to be best practice, but deemed to have limited feasibility 
Multidisciplinary teams (consultants only) 
Professionals who are knowledgeable about adolescent development 
Age-appropriate physical environment 
Dedicated adolescent environments (e.g. adolescent waiting areas) 
Providing opportunities to meet similar others 

Effect of the Delphi technique on agreement and consensus

The Delphi technique had only a limited effect on agreement. No significant changes in agreement for best practice were observed between Rounds 1 (data not shown) and 2 (Table 2) for any of the groups. Agreement regarding feasibility increased for 2 (8%) items and declined for 3 (12%). This may have been influenced by the fact that providers wrote many more comments in response to the feasibility of Round-1 items than they did for items of best practice.

The effect of the Delphi technique on the consensus within the groups was similarly limited. Consensus regarding best practice significantly increased in 3 (12%) items and increased in 5 (19%) items regarding feasibility.

Discussion

This study endeavoured to define ‘best practice’ and ‘feasibility’ in the transitional care of adolescents with JIA and by integrating the views of users and providers is likely to have produced findings that are extremely salient for future developments in this area. In view of the generic nature of the main findings reported here, this study has potential wider relevance to all young people with chronic illness requiring transitional care.

Triangulation of the findings with the previous needs assessment [15, 16] and existing policy statements [1–3] shows a high degree of concordance, and further supports the validity of the findings. Perhaps most importantly, this study provides useful directions for future initiatives; those components deemed important and feasible should be implemented now (Table 4), whereas those rated as important but not feasible (Table 5) should become priorities for subsequent research, development and service commissioning.

Areas of transitional care to be implemented now

Young people's psychosocial and educational/vocational needs

All groups felt that it was important to address young people's psychosocial and educational/vocational needs (Table 2). This is a central tenet of transitional care [1–3] and is important when one considers that young people with JIA are at risk of unemployment [4, 8] and relationship difficulties [6] compared with controls. Adolescents may have more diverse concerns than health professionals anticipate [17, 18]. However, whilst providers felt it was feasible to address these issues in most, if not all, hospitals, it is likely that specific training will be needed (e.g. knowledge of appropriate referral routes) [3, 19].

An individualized approach

Transition moves at different rates for different people and therefore requires the development of individual transition plans that are developmentally appropriate [1, 20]. The results of this study support these sentiments and echo current proposals [1, 3]. Furthermore, such planning should be anticipatory and be regularly reviewed (Table 2).

An individualized approach may also depend upon key workers to coordinate the planned activities [21] and facilitate inter-agency communication. All groups agreed this constituted best practice, AHPs more strongly than consultants (Table 3). The key worker role has been postulated to be best fulfilled by a nurse specialist [2], although other AHPs may be equally well qualified.

Honest explanations of the adolescent's condition and associated health-care

Several studies report that young people feel that neither parents nor professionals should act as gatekeepers in the disclosure of negative news (e.g. adverse side-effects) [16, 22–24]. Barriers to the effective communication of information include interpersonal factors (e.g. lack of comfort with adolescent issues, lack of perceived applicability, ambivalence with respect to role), communication skills of the adolescent and health professional, and structural issues (not routinely seeing adolescent alone, duration and frequency of contact) [24, 25]. Effective communication skills can be learned [26] but need to address a range of practical, attitudinal and behavioural factors [24].

Opportunities for adolescents to express opinions and make informed decisions

The importance of providing opportunities for adolescents to express opinions and make informed decisions is in keeping with guidelines [3, 20] and various legislation [27, 28]. Young people are often unsure of their rights to confidentiality and to accept or decline treatment [16]. Professionals involved with this age-group need to be aware of these issues and to educate young people and their parent(s) accordingly, including the facilitation of the development of self-advocacy skills (Table 2). Related to this are opportunities for ‘free and open access to all available information regarding JIA and related health issues’, which was agreed by all groups to be best practice and feasible in most hospitals, again echoing national recommendations [1, 2]. However, in practice, this is not without practical challenges (e.g. provision of sexual health information for adolescents in the out-patient area, where all ages of children are seen). Moreover, paediatricians who are used to younger patients may not appreciate that young adolescents may have specific health concerns and have already established patterns of risky behaviour [29, 30].

Continuity in health personnel

This call for continuity is seen as a key factor in building trust [16, 24] and is particularly pertinent to the developmental stage of this age group as they gradually gain confidence to independently negotiate the health-care system. Continuity is also viewed by young people and their parents as an important prerequisite for the parents’ eventual withdrawal from their child's consultations [16]. This issue needs to be acknowledged in the development of future rheumatology services for adolescents.

Give adolescents the option of being seen by professionals without their parents

Giving adolescents the option of being seen without their parents was deemed important and feasible by all groups. Of note, however, is that young people agreed with this more strongly than parents (P = 0.003). Discrepancies between parents’ and health providers’ opinions in the age for commencement of independent visits have been reported, parents reporting a later start for certain independent behaviours [31]. Only a minority of young people are seen independently in rheumatology clinics [10, 25, 32]. In stark contrast, at least half of their healthy peers are seeing their general practitioner alone by the age of 14–15 yr [33]. Independent visits facilitate confidentiality for the young person, which, in primary care, has been reported as a key attribute of adolescent-friendly practice [34]. In one US study, paediatric practices were less likely than family medicine practices to offer confidential services to adolescents [35].

One of the ways to support adolescents’ independence may be to offer concurrent visits for parents, a concept which was rated higher by users (both patients and parents) than by consultants (Table 3). Parents often have their own transitional care needs and want to talk to members of the team in confidence [16]. Family connectedness is an important factor in the resilience of young people with chronic illness and/or disability [36] and, therefore, an integral component of transitional care. However, whilst AHPs felt concurrent visits to be feasible, consultants felt this was attainable in only some hospitals, perhaps reflecting geographical disparities in multidisciplinary team-working. Concurrent visits have major implications for service provision in terms of staffing and overall consultation time per patient. Such requirements must be communicated to the commissioning authorities if appropriate adolescent rheumatology services are to be developed in the future.

Targets for future research and development

Multidisciplinary teams (consultants only)

The difference in providers regarding the feasibility of transitional care being provided by multidisciplinary teams may reflect AHPs’ experience of multidisciplinary working. This may also explain why AHPs typically exhibited the greatest strength of agreement regarding best practice. For some rheumatologists, the multidisciplinary team is a virtual one, with links to the AHPs made only via paper referrals. The difference in agreement may also represent variations in the understanding and belief systems about transition. This, in part, is understandable in view of the lack of developments in adolescent rheumatology and transitional care in the UK [10], particularly in respect to training [2, 3, 19]. Future developments in transitional care training must acknowledge the multidisciplinary nature of transition.

Professionals who are knowledgeable about adolescent development

The lack of relevant training [19] may also explain why providers reported limited feasibility of transitional care being provided by ‘professionals who are knowledgeable about adolescent development’. Accordingly, the recommendation that adolescent health should be included in both undergraduate and postgraduate curriculums is welcomed [2].

Age-appropriate physical environment and dedicated adolescent environments (e.g. adolescent waiting areas)

Age-appropriate and dedicated adolescent environments were considered by health professionals to constitute best practice but of limited feasibility. Such facilities are currently underdeveloped in UK hospitals [2, 10, 37] and have been called for in the children's NSF [1].

Whilst there are significant barriers to specialized adolescent facilities (fiscal and environmental), improvement of the current situation may require only small changes. When young people in general practice were questioned about the attributes of an adolescent-friendly primary health-care service, the top six attributes were confidentiality, phone advice, written information, a special clinic, friendliness and magazines [34]. Most of these are potentially feasible in adult or paediatric hospitals at minimal cost.

Proving opportunities to meet similar others

Providing opportunities to meet similar others was also reported to be of limited feasibility. This said, the clinic waiting area is a simple, albeit underused, arena for providing such opportunities. Introduction to similar others facilitated by professionals was suggested by young people in a previous needs assessment [16]. Some units have also met this need by residential breaks for teenagers, funded by local charities [38].

Limitations

A number of caveats should be borne in mind. First, ‘findings of a Delphi represent expert opinion, rather than indisputable fact’ [39, p. 382]. This said, in a relatively new area of health-care with a minimal evidence base, it is imperative to consider such opinion. Secondly, specialists tend to favour interventions with which they are most familiar. Lack of familiarity with the principles and philosophy of transition, against a background of reported training needs in this area [19], may therefore have biased the findings.

With respect to providers, this study did not reflect the full spectrum of potential key stakeholders in adolescent rheumatology transitional care, e.g. primary health-care professionals, education, social services and voluntary agencies. It is reassuring, therefore, that multidisciplinary working, interagency links and educational and vocational issues were supported by the AHPs and/or the consultants as being best practice and/or feasible. Consensus-based guidelines should therefore be interpreted in the context of the speciality composition of the group [12].

Users frequently had the lowest median scores in rating best practice. This may reflect the fact that each user score represented a single familial pairing and hence limited experience of adolescent rheumatology health-care, particularly transitional care. This is compared with the numerous families seen by the providers in their day-to-day work, who may have experience of working in different and geographically diverse settings. It is also possible that the similarity of patient–parent responses may be attributable to a lack of independence in the completion of questionnaires. However, whilst independence cannot be guaranteed, the significant difference between patients’ and parents’ scores with respect to young people being seen on their own in clinic would support a degree of individual completion.

Using pre-existing information in the development of the questionnaire may also have biased the responses or limited the available options [40]. Moreover, the results may have been influenced by ambiguity in the phrasing of items. For instance, in response to the item ‘Adolescents and parents need early preparation for their transfer to adult rheumatology services (e.g. starting at 11 years)’ several panel members commented that 11 years was too young. This item was agreed to constitute best practice, but consensus was lower than for many of the other items. In part, this may have been attributable to the level of specificity in the example age given, and greater validity may have been achieved with a more comprehensive pilot study.

Overall, the impact of the Delphi technique on agreement and consensus appeared greatest where feedback was greatest. Thus, future work may benefit by using face-to-face methods that encourage social interaction and give opportunities to probe discordant opinions.

Summary

In summary, this study has defined best practice in transitional care for adolescents with JIA, as perceived by users and providers, and has examined the feasibility of achieving this within a UK National Health Service context. By doing so, it is hoped that the components of best practice identified will be supported by the key clinical groups involved and taken up in clinical practice and policy. Statements seen as best practice and feasible should be implemented now. Statements seen as best practice but not feasible should be the focus of future research and development.

This work was funded by a Special Purpose Grant (SO635) from the Arthritis Research Campaign (www.arc.org.uk) and undertaken on behalf of the British Paediatric Rheumatology Group, the Children's Chronic Arthritis Association (www.ccaa.org.uk), the Lady Hoare Trust for physically disabled children (www.ladyhoaretrust.org.uk) and Arthritis Care (www.arthritiscare.org.uk). KLS coordinated the project, co-designed and distributed the questionnaires, performed the data analyses and co-wrote the paper. JEMcD is an Arthritis Research Campaign Clinical Senior Lecturer in Paediatric Rheumatology. She co-conceived the study, wrote the grant application, obtained funding, directed the project, co-designed the questionnaire and co-wrote the paper. TRS co-conceived the study, supervised the grant application and reviewed the paper prior to submission. Dr Carole Cummins helped review the provisional draft. Sincere thanks are extended to all the organizations and individuals that participated in the study. Thanks are also offered to Mrs Bev Thomas, who helped input the data.

The authors have declared no conflicts of interest.

References

1
Department of Health. Getting the right start: national service framework for children. Standard for Hospital Services. April 2003 (www.doh.gov.uk/nsf/).
2
Royal College of Paediatrics and Child Health. Bridging the gap: health care for adolescents. June 2003 (www.rcpch.ac.uk).
3
American Academy of Pediatrics, American Academy of Family Physicians, American College of Physicians–American Society of Internal Medicine. A consensus statement on health care transitions for young adults with special health care needs.
Pediatrics
 
2002
;
110
:
1304
–6.
4
Oen K, Malleson PN, Cabral DA, Rosenberg AM, Petty RE, Cheang M. Disease course and outcome of juvenile rheumatoid arthritis in a multicentre cohort.
J Rheumatol
 
2002
;
29
:
1989
–99.
5
Foster HE, Marshall N, Myers A, Dunkley P, Griffiths ID. Outcome in adults with juvenile idiopathic arthritis. A quality of life study.
Arthritis Rheum
 
2003
;
48
:
767
–75.
6
Packham JC, Hall MA. Long-term follow-up of 246 adults with juvenile idiopathic arthritis: social function, relationships and sexual activity.
Rheumatology
 
2002
;
41
:
1440
–3.
7
Packham JC, Hall MA. Long-term follow-up of 246 adults with juvenile idiopathic arthritis: functional outcome.
Rheumatol
 
2002
;
4
:
1428
–35.
8
Packham JC, Hall MA. Long-term follow-up of 246 adults with juvenile idiopathic arthritis: education and employment.
Rheumatology
 
2002
;
4
:
1436
–9.
9
Forbes A, While A, Ullman R et al. A multi-method review to identify components of practice which may promote continuity in the transition from child to adult care for young people with chronic illness or disability. London: National Co-ordinating Centre for NHS Service Delivery and Organisation R & D,
2001
.
10
McDonagh JE, Foster HE, Hall MA, Chamberlain MA. Audit of rheumatology services for adolescents and young adults in the UK.
Rheumatology
 
2000
;
39
:
596
–602.
11
Vella K, Goldfrad C, Rowan K, Bion J, Black N. Use of consensus development to establish national research priorities in critical care.
BMJ
 
2000
;
320
:
976
–80.
12
Murphy MK, Black NA, Lamping DL et al. Consensus development methods, and their use in clinical guideline development.
Health Technol Assessment
 
1998
;
2
:
1
–88.
13
Jones J, Hunter D. Qualitative research: consensus methods for medical and health services research.
BMJ
 
1995
;
311
:
376
–80.
14
Ayanian JZ, Landrum, MB, Normand ST, Guadagnoli E, McNeil BJ. Rating the appropriateness of coronary angiography—do practicing physicians agree with an expert panel and with each other?
N Engl J Med
 
1998
;
338
:
1896
–904.
15
Shaw KL, Southwood TR, McDonagh JE. User perspectives of transitional care for adolescents with juvenile idiopathic arthritis.
Rheumatology Advance Access published March 23
 ,
2004
, 10.1093/rheumatology/keh175.
16
Shaw KL, Southwood TR, McDonagh JE. Identifying the needs of adolescent with juvenile idiopathic arthritis: results of nationwide focus groups.
Rheumatology
 
2004
(in press)
.
17
Carroll G, Massarelli E, Opzoomer A et al. Adolescents with chronic disease: are they receiving comprehensive health care?
J Adolesc Health Care
 
1983
;
17
:
32
–6.
18
Kowpak M. Adolescent health concerns: a comparison of adolescent and health care provider perceptions.
J Am Acad Nurse Pract
 
1991
;
3
:
122
–8.
19
McDonagh JE, Southwood TR, Shaw KL. Unmet education and training needs of rheumatology health professionals in adolescent health and transitional care.
Rheumatology Advance Access published March 2
 ,
2004
, 10.1093/rheumatology/keh163.
20
Reiss JR, Gibson R. Health care transition: destinations unknown.
Pediatrics
 
2002
;
110
:
1307
–14.
21
Garwick AW, Kohrman C, Wolman C, Blum RW. Families’ recommendations for improving services with chronic conditions.
Arch Pediatr Adolesc Med
 
1998
;
152
:
440
–8.
22
Ullrich G, Mattussek S, Dressler F, Thon A. How do adolescents with juvenile chronic arthritis consider their disease related knowledge, their unmet service needs and the attractiveness of various services?
Eur J Med Res
 
2002
;
7
:
8
–18.
23
Barlow JH, Shaw KL, Harrison K. Consulting the ‘experts’: children's and parents’ perceptions of psycho-educational interventions in the context of juvenile chronic arthritis.
Health Educ Res
 
1999
;
14
:
597
–610.
24
Beresford BA, Sloper P. Chronically ill adolescents’ experiences of communicating with doctors: a qualitative study.
J Adolesc Health
 
2003
;
33
:
172
–9.
25
Britto MT, Rosenthal SL, Taylor J, Passo MH. Improving rheumatologists’ screening for alcohol use and sexual activity.
Arch Pediatr Adolesc Med
 
2000
;
154
:
478
–83.
26
Maguire P, Pitceathly C. Key communication skills and how to acquire them.
BMJ
 
2002
;
325
:
697
–700.
27
Children's Act. London: HMSO,
1989
.
28
United Nations. Convention on the rights of the child. Geneva: United Nations,
1989
.
29
Nash AA, Britto MT, Lovell DJ, Passo MH, Rosenthal SL. Substance use among adolescents with juvenile rheumatoid arthritis.
Arthritis Care Res
 
1998
;
11
:
391
–6.
30
Millstein SG, Irwin CE, Adler NE et al. Health risk behaviours and health concerns among young adolescents.
Pediatrics
 
1992
;
89
:
422
–8.
31
Geenen SJ, Powers LE, Sells W. Understanding the role of health care providers during the transition of adolescents with disabilities and special health care needs.
J Adolesc Health
 
2003
;
32
:
225
–33.
32
Robertson LP, Shaw KL, Southwood TR, McDonagh JE. An audit of patients with juvenile idiopathic arthritis recently transferred from paediatric to adult rheumatology clinics.
Rheumatology
 
2003
;
42(Suppl. 1)
:
51
.
33
Jacobson L, Owen P. Study of teenage care in one general practice.
Br J Gen Pract
 
1993
;
43
:
349
.
34
McPherson A. Primary health care and adolescence. In: MacFarlane A, ed. Adolescent medicine. London: Royal College of Physicians,
1996
:
33
–41.
35
Akinbami LJ, Gandhi H, Cheng TL. Availability of adolescent health services and confidentiality in primary care practices.
Pediatrics
 
2003
;
111
:
394
–401.
36
Patterson J, Blum RJ. Risk and resilience among children and youth with disabilities.
Arch Pediatr Adolesc Med
 
1996
;
150
:
692
–98.
37
Suresh S, Doull IJM, Thomas P. Adolescent inpatients units.
Arch Dis Child
 
2000
;
82
:
266
.
38
Hackett J, Johnson B. Evaluation of an independence break for teenagers with rheumatic disease.
Ann Rheum Dis
 
2001
;
60(Suppl. 52)
:
198
.
39
Powell C. The Delphi technique: myths and realities.
J Adv Nurs
 
2003
;
41
:
376
–82.
40
Hasson F, Keeney S, McKenna H. Research guidelines for the Delphi survey technique.
J Adv Nurs
 
2000
;
32
:
1008
–15.

Comments

0 Comments