Self-care support for children and adolescents with long-term conditions: the REfOCUS evidence synthesis

Population

We defined children and young people as individuals aged < 18 years. Although the transition to adult services is not always immediate and key elements of development may continue beyond 18 years of age, this cut-off point aligned with our earlier reviews on the clinical effectiveness of self-care support interventions for children and young people. In accordance with the inclusion criteria of our previous reviews, we included studies with participants aged up to 25 years as long as the mean age of the sample, and/or the majority of participants, remained under the age of 18 years.

We restricted our review to LTCs. To be eligible for inclusion in the review, participants were required to have a diagnosis of a LTC, defined through clinical assessment, contact with health services or symptom scores above clinical cut-off points on validated screening instruments. We excluded preventative studies that looked at a population at ‘high risk’ of developing a LTC.

There is no definitive list of LTCs and hence we adopted the Department of Health’s generic definition of a LTC as one ‘that cannot be cured but can be managed through medication and/or therapy’. 11 We included studies recruiting patients with a mix of LTCs.

Both mental and physical health conditions were eligible for inclusion in the review. This included common conditions such as diabetes, asthma, coronary heart disease, depression, anxiety and psychosis. Comprehensive lists of eligible conditions are provided in Box 2. In line with the views of our patient and public involvement (PPI) advisory panel, we excluded autism spectrum disorder, intellectual disabilities, substance misuse (unless comorbid with another LTC) and cancer in long-term recovery or remission, as these conditions were deemed to fall outside our working definition of a long-term physical or mental health condition.

Interventions

Self-care can be defined in different ways according to who engages in the self-care behaviour (e.g. individual, family, community) and the intervention context (e.g. health promotion, illness prevention, illness impact limitation or restoration of health). To meet the definition of self-care support, an intervention needs to include an agent other than the self, typically a health professional, peer group, voluntary sector representative or information technology platform.

The goal of self-care support has previously been defined as the enablement of patients to perform three discrete sets of tasks: medical management of their condition (e.g. taking medication); carrying out normal roles and activities; and managing the emotional impact of their condition. 64 For the purposes of our review, we defined a self-care support intervention as:

. . . any intervention primarily designed to develop the abilities of children and young people (and/or their adult carers) to undertake management of their long-term health condition through education, training and support to develop their knowledge, skills or psychological and social resources.

Example categories of self-care support of relevance to this review are outlined in Box 3. We included all formats and delivery methods for self-care support (e.g. group or individual, face to face or remote, professional or peer led). Interventions delivered in health, social care, educational or community settings were included. Interventions that targeted the child or young person, or their adult caregiver, were included.

Written action plans, developed in collaboration with children and young people or their families, were eligible for the review, but were excluded if there was no evidence of self-care discussion or negotiation. Self-care support, by definition, is designed to offer a more participatory approach to health care, with patients making a critical contribution to achieving health gain and making decisions to ensure that their care is personalised to their needs. We excluded all interventions where the target of the intervention was not actively engaged and/or remained a passive recipient of knowledge or instructions.

We excluded self-care undertaken without any input, guidance or facilitation by services. Although self-care can be, and often is, undertaken without service support, it is rarely the subject of intervention studies. We excluded studies where the effects of self-care support could not be distinguished from broader interventions for LTCs. We excluded studies evaluating service development or quality improvement initiatives in which self-care support was not the predominant component of the intervention.

Comparators

We included studies in which a self-care support intervention was additional to usual care and compared against usual care alone, or in which a self-care support intervention was compared against a more intensive usual care intervention (e.g. home- vs. clinic-based monitoring). We excluded studies in which two versions of self-care support were compared and the two interventions were of comparable intensity and content, because such comparisons did not allow for an assessment of the impact of self-care support per se.

Outcomes

To meet our research objectives, we required evidence of effectiveness of validated self-care support to reduce health-care utilisation without compromising children and young people’s outcomes. We restricted our analysis to studies of self-care support that reported quantitative data on patient outcomes and health-care utilisation, as these were the only studies that could answer our brief.

Eligible patient outcomes included standardised measures of health-related or generic QoL or disease-specific symptom measures or events. We excluded intermediate outcomes and measures of psychological or clinical variables that did not provide an assessment of subjective health status or QoL [e.g. self-care behaviours, self-efficacy, glycated haemoglobin (HbA_1C) levels or forced expiratory volume recordings]. In adult populations, such variables are known to be unreliable indicators of health-related quality of life (HRQoL). 65 We extracted data on the health outcomes of the child/young person and excluded the health outcomes of adult caregivers.

Eligible outcomes for health-care utilisation comprised data on hospital visits and admissions, emergency care, primary care visits, other scheduled or unscheduled health-care use, patient costs and total costs. Our primary foci were comprehensive measures of health service costs (i.e. summed totals of multiple sources of cost) and/or major cost drivers (i.e. hospital admissions). Other, more minor, costs (such as medication use) were identified but not formally analysed. The rationale for this is discussed further in Data preparation and analysis.

Design

We included randomised controlled trials (RCTs), non-randomised controlled trials (nRCTs), controlled before-and-after studies (CBAs) and interrupted time series designs, as defined according to the Effective Practice and Organisation of Care (EPOC) criteria63 (Box 4). UK and non-UK studies were included. Translation of non-English-language studies was undertaken.

Electronic databases

We began the process of identifying eligible studies by checking published reviews, including those previously undertaken by the research team. 26,31,32 We complemented our searches of existing reviews with a primary search of multiple electronic databases, conducted in March 2015.

We updated and expanded our existing search strategies to ensure that they were sensitive to a broad range of health-care utilisation beyond formal cost-effectiveness analyses. Search terms relating to the key concepts of the review were identified by scanning the background literature and browsing the MEDLINE medical subject heading thesaurus, and through discussion with collaborating colleagues at the University of York’s Centre for Reviews and Dissemination.

A search strategy was developed in MEDLINE, using an iterative approach tested against a set of 15 studies known to be relevant to our review. This MEDLINE search strategy was adapted to run on all other databases designated in our protocol.

Electronic searches were undertaken on the following health and allied health databases:

• MEDLINE (accessed 18 March 2015 via OvidSP; www.ovissp.ovid.com)

• EMBASE (accessed 18 March 2015 via OvidSP; www.ovissp.ovid.com)

• PsycINFO (accessed 17 March 2015 via OvidSP; www.ovissp.ovid.com)

• Cumulative Index to Nursing and Allied Health Literature (CINAHL; accessed 19 March 2015 via EBSCOhost; www.search.ebscohost.com)

• ISI Web of Science, including Social Sciences Citation Index (SSCI) and Science Citation Index Expanded (accessed 19 March 2015 via Web of Science; www.wos.mimas.ac.uk)

• NHS Economic Evaluation Database (NHS EED) (accessed 18 March 2015 via Wiley Online Library)

• The Cochrane Library, including Cochrane Database of Systematic Reviews (CDSR), Database of Abstracts of Reviews of Effects (DARE) and Cochrane Central Register of Controlled Trials (CENTRAL) (accessed 18 March 2015 via Wiley Online Library)

• Health Technology Assessment database (accessed 18 March 2015 via Wiley Online Library)

• Paediatric Economic Database Evaluation (PEDE) (accessed 31 March 2015 via http://pede.ccb.sickkids.ca/pede/index.jsp)

• the IDEAS database of economic and finance research (accessed 31 March 2015 via http://ideas.repec.org).

All databases were searched from inception. No language or design restrictions were applied. Full details of the search strategies, search terms and the specific dates of individual searches are reported in Appendix 1.

Additional search strategies included scanning the bibliographies of all relevant retrieved articles, targeted author searches (for additional publications and/or unpublished data identified in conference abstracts) and forward citation searching. No studies were identified that had not been retrieved by other means.

Changes to the search protocol

All searches were conducted as specified in the original review protocol with the exception of the Health Economic Evaluations Database (HEED). HEED ceased publication prior to study commencement and was not searched as part of the final review. Coverage of the relevant economic evidence base was ensured through searches of the NHS EED, the Health Technology Assessment database, the PEDE and the IDEAS database of economic and finance research. The potential impact of this protocol change was judged to be minimal.

Methodological quality appraisal

Methodological quality appraisals were undertaken by one reviewer and independently verified by a second reviewer. Studies were assessed for methodological quality using the Cochrane Collaboration Risk of Bias Assessment Tool for RCTs62 or the Cochrane guidance for non-randomised designs. 62 Economic studies were assessed using a critical appraisal checklist for economic evaluations. This checklist was based on Drummond’s checklist for assessing economic evaluations67 and was adapted to capture more fully the quality of economic evaluations in self-care support interventions (see Appendix 2).

Quality ratings for randomised studies were based on a dichotomous measure of allocation concealment (i.e. adequate or inadequate/unclear). Allocation concealment is the aspect of trial quality most consistently associated with treatment effect. 68,69 Other indicators may be less relevant in trials of behavioural interventions where participant, personnel and outcome blinding are often difficult to achieve.

Small study bias

Funnel plots74 using standard errors75 and associated regression tests were used to explore small-study bias where sufficient data were available. The purpose of a funnel plot is to map standardised ESs from individual studies against their standard error (i.e. the underlying precision of the observed effect). A funnel plot is based on the premise that precision in an ES estimate will increase as sample size increases. Effect estimates from smaller studies with larger standard errors should, therefore, scatter more widely at the bottom of the plot. Larger studies with smaller standard error should display a narrower spread. Bias is suggested by an asymmetrical plot and statistical testing of a potential relationship between treatment effect and precision. An absence of smaller studies without statistically significant effects is an indicator of potential publication bias. In this situation, the effect calculated in a meta-analysis may overestimate the intervention effect.

Changes to the analytical protocol

Our analysis was designed to consider the ability of models of self-care to reduce health-care costs without compromising patient outcomes. Our primary analysis was on total costs. Our protocol stipulated that our secondary analyses would, where data allowed, consider all other major types of resource use and cost. This included inpatient, outpatient, primary care, community care and patient out-of-pocket expenditures.

Meaningful analysis requires that sufficient, comparable data are reported across the primary studies. Lack of consistent measurement and ambiguity in some of the outcomes that were reported prevented accurate demarcation of primary, secondary and community health-care costs. More usually, outcome data were presented as urgent (non-scheduled) compared with scheduled service use. Definitions of scheduled resource use varied according to illness type and context.

Our PPI advisory panel identified hospital admissions, ED visits and patient and families’ out-of-pocket expenses as the three outcomes that they would like to be prioritised in our review. An insufficient number of studies reported out-of-pocket expenses. Our secondary analyses thus focused on hospital admissions and ED use.

Hospital use represents a significant driver of total costs in most health-care systems. However, focusing on a single source of utilisation leaves the analysis vulnerable to cost shifting, where any benefits found in terms of reduced hospital use may mask increased costs elsewhere in the health-care system (such as in community care). Our primary analysis thus remained focused on total costs.

Post hoc classification by long-term condition

We grouped different LTCs post hoc into four conceptually and clinically relevant categories. These categories were asthma, other (non-asthma) physical health conditions, behavioural disorders and mental health.

Our a priori intention was to also aggregate data across subtypes or ‘clusters’ of conditions, based on a similar typology to that developed by the Practical systematic RevIew of Self-Management Support for long-term conditions (PRISMS) study for adults with LTCs. 26 We did not aggregate our data in this way, as all but four studies focusing on behavioural disorders fell into the same condition cluster (cluster 1: LTCs with marked variability in symptoms over time).

Post hoc classification by intervention type

Existing typologies of self-care support for children and young people with LTCs highlight the importance of considering different aspects and characteristics of the intervention, including its target, location, facilitation and delivery methods. 31

We conducted subgroup analyses based on intervention target (child and/or young person, adult or both), format (individual, group or mixed) and delivery method (face to face, remote or mixed model). We also conducted subgroup analyses on intervention setting, defined as inpatient, outpatient/clinic, school or community, home or mixed location.

In line with our previous review of self-care support for adults with LTCs,26 we included interventions across the spectrum of care and distinguished post hoc between the different intensities and types of self-care support that were provided.

We used a similar approach to classify intervention intensity as we used in our previous review, with post hoc amendments to accommodate the level and type of intervention descriptions provided in our primary studies. Our final classification system was informed and approved by our PPI advisory panels and distinguished between four different categories of self-care support:

1. ‘Pure’ self-care support for interventions providing self-care support through a stand-alone resource (e.g. interactive mobile application or educational online program).

2. Facilitated self-care support for interventions providing fewer than four sessions or < 2 hours of face-to-face or remote self-care support. Support is provided by a designated self-care agent (e.g. health professional or peer) and usually targets a single group (e.g. children or parents). The support provided will often be (but is not limited to) self-care education, feedback or care plan review.

3. Intensively facilitated self-care support for interventions providing regular and repeated contact exceeding more than four sessions or 2 hours’ support in total. Support is provided by a designated self-care agent health professional or peer and often targets multiple groups (e.g. children and parents or children and teachers). The support provided will often be multifaceted and may include some co-ordination of a patient’s primary or standard care.

4. Case management for interventions providing more than four sessions or 2 hours of additional support from a designated agent, with additional support from a multidisciplinary team and explicit referrals or care co-ordination as part of the intervention protocol.

Two authors independently assessed the type, and content, of each self-care support intervention. Disagreements were identified and resolved via team discussion.

Primary analysis: quality of life and total health service costs

Total health service costs were infrequently reported. Only eight studies reporting 10 comparisons were eligible for inclusion in a permutation plot that simultaneously charted the effects of self-care support on children and young people QoL and total health-care costs (Figure 9). Six of these comparisons were rated as being at a low risk of bias.

FIGURE 9.

Permutation plot of QoL and costs.

When effects were plotted against each other, the comparisons were primarily distributed across the left-hand quadrants of the plot, suggesting that self-care support interventions currently demonstrate high variability in terms of economic effect, but typically confer minimal to small improvements for QoL. This conclusion is based on limited data and must be treated with caution. The circles in the permutation plots are an illustrative indicator of their relative ‘weight’ in the analysis. Permutation plots do not consider uncertainty around individual study point estimates which, in some instances, may be marked. Almost all studies reporting total costs (eight comparisons) demonstrated significant skew in either control or intervention outcome data.

Quality of life and hospital admissions

Fifty-three comparisons were eligible for inclusion in a permutation plot charting the effects of self-care support on QoL and hospital admissions (Figure 10); 29 of these comparisons originated from RCTs with adequate allocation concealment.

FIGURE 10.

Permutation plot of QoL and hospital admissions.

When hospital admissions were plotted against patient outcomes, most comparisons were distributed on the left-hand side, spanning both the lower and upper left-hand quadrants. This suggests that, on the basis of the available evidence, self-care support for children and young people is likely to be associated with improvements in QoL, but variable effects on hospital admissions. A minority of studies was located in the lower right-hand quadrant, suggesting reduced hospital admissions, but a marginally compromised QoL. As stated previously, permutation plots do not consider the magnitude of uncertainty around individual study point estimates and, for some studies in the current analysis, this uncertainty may be marked.

Quality of life and emergency department visits

Emergency department visits were identified by our PPI panel as a particularly important aspect of health service utilisation for children, young people and their parents. Forty-seven comparisons were eligible for inclusion in this permutation plot (Figure 11); 24 were from RCTs with adequate allocation concealment.

FIGURE 11.

Permutation plot of QoL and emergency visits.

When emergency visits were plotted against patient outcomes, the majority of studies fell in the lower left-hand quadrant, demonstrating that self-care support can reduce ED use without routinely compromising children and young people’s QoL. Fewer studies report reduced emergency visits with decrements in QoL (lower right-hand quadrant) or significant improvements in QoL associated with increased service use (upper left-hand quadrant).

Asthma

Sixty-six studies evaluated self-care support for children and young people with asthma. The flow of studies through the review is depicted in Figure 12. Pooled effects for each outcome are reported in Table 3. Meta-analysis of all asthma studies demonstrated that self-care support was associated with minimal but statistically significant improvements in QoL, with moderate variation across trials. Self-care support was associated with minimal but statistically significant reductions in ED use, with low variation across the studies. Meta-analyses showed no significant effects on hospital admissions. Meaningful interpretation of total cost data was limited by the small number of comparisons (n = 2).

FIGURE 12.

Analyses of studies for patients with asthma. CYP, children and young people.

Owing to a lack of data, permutation plots were not calculated for total costs. Thirty-eight comparisons were eligible for inclusion in a permutation plot charting the effects of self-care support on QoL and hospital admissions for asthma (Figure 13); 16 of these comparisons originated from RCTs with adequate allocation concealment.

FIGURE 13.

Permutation plot of QoL and hospital admissions (asthma).

When hospital admissions were plotted against patient outcomes, most comparisons were distributed across the lower right- and left-hand quadrants. This suggests that self-care support interventions that reduce the number of hospital admissions for children and young people with asthma will not routinely compromise QoL but, on the basis of the current evidence, such compromises cannot be ruled out.

When emergency visits were plotted against QoL for children and young people with asthma (Figure 14), the majority of studies fell in lower left-hand quadrant, demonstrating that self-care support can reduce ED use without compromising children and young people’s QoL. A notable number of studies in other quadrants suggested that self-care support interventions may reduce emergency visits with decrements in QoL (lower right-hand quadrant) or improve in QoL but increase service use (upper left-hand quadrant).

FIGURE 14.

Permutation plot of QoL and emergency visits (asthma).

Other (non-asthma) physical health conditions

Eight studies evaluated self-care support for children and young people with other physical health conditions. The flow of studies through the review is depicted in Figure 15. Owing to the small number of data available for meta-analysis, meaningful interpretation of the evidence base for non-asthma physical health conditions is limited. Pooled ESs are presented in Table 4 for completeness. Permutation plots are not presented.

FIGURE 15.

Analyses of studies for patients with other (non-asthma) physical health conditions. CYP, children and young people.

Mental health conditions

Eighteen studies evaluated self-care support for children and young people with mental health conditions. The flow of studies through the review is depicted in Figure 16. Pooled effects for each outcome are reported in Table 5. Meta-analysis of all mental health studies demonstrated that self-care support was associated with minimal but statistically significant improvements in QoL, with moderate variation across trials. The meta-analyses showed no significant effects on hospital admissions, ED visits or total costs. Meaningful interpretation of total cost data was limited by a small number of comparisons (n = 8) and high variation across trials.

FIGURE 16.

Analyses of studies for patients with mental health conditions. CYP, children and young people.

Owing to a lack of data, permutation plots were not calculated for total costs. Fourteen comparisons were eligible for inclusion in a permutation plot charting the effects of self-care support on QoL and hospital admissions for mental health (Figure 17); 10 of these comparisons originated from RCTs with adequate allocation concealment.

FIGURE 17.

Permutation plot of QoL and hospital admissions (mental health conditions).

When hospital admissions were plotted against patient outcomes, the majority of comparisons were located in the lower left-hand quadrant, suggesting that self-care support can reduce utilisation for children and young people with mental health conditions without compromising QoL. A minority of studies were located in the lower right-hand quadrant, suggesting reduced hospital admissions but a marginally compromised QoL. As stated previously, data were limited and findings must be treated with caution.

Nine comparisons were eligible for inclusion in a permutation plot charting ED visits against patient outcomes (Figure 18); seven were from RCTs with adequate allocation concealment. When emergency visits were plotted against patient outcomes, the majority of studies fell in lower left-hand quadrant, demonstrating that self-care support can reduce ED use without routinely compromising children and young people’s QoL. Limited data mean that these results must be treated with caution.

FIGURE 18.

Permutation plot of QoL and emergency visits (mental health conditions).

Behavioural difficulties

Five studies evaluated self-care support for children and young people with behavioural difficulties. The flow of studies through the review is depicted in Figure 19.

FIGURE 19.

Analyses of studies for patients with behavioural difficulties. CYP, children and young people.

Owing to the small number of data available for meta-analysis, meaningful interpretation of the evidence base for non-asthma physical health conditions is limited. Pooled ESs are presented in Table 6 for completeness. Permutation plots are not presented. Table 7 summarises the results of all meta-analyses, presented according to LTC type.

Small-study bias

The funnel plots for QoL and health utilisation outcomes are presented in Figures 20–23. A funnel plot is based on the premise that precision in the estimation of an ES will increase as sample size increases. Bias is suggested by the emergence of a non-symmetrical plot.

FIGURE 20.

Funnel plot: QoL. ESQoL, effect size QoL; SE, standard error.

FIGURE 21.

Funnel plot: hospital admissions. ESADM, effect size hospital admissions.

FIGURE 22.

Funnel plot: emergency visits. ESEmerg, effect size emergency visits.

FIGURE 23.

Funnel plot: total costs. ESTot, effect size total costs.

Potential for publication bias in QoL outcomes was observed (Egger’s bias 0.79, 95% CI 0.08 to 1.51; p = 0.03). This result is most likely influenced by a single study on the bottom left-hand side of the funnel plot. No evidence of publication bias was observed for hospital admissions (Egger’s bias 0.16, 95% CI –0.82 to 1.14; p = 0.74), emergency visits (Egger’s bias –0.19, 95% CI –1.00 to 0.62; p = 0.64) or total costs (Egger’s bias –3.40, 95% CI –15.25 to 8.46; p = 0.53), although the low power of this final test means that publication bias cannot definitely be ruled out.

Reach

The reach of health behaviour interventions refers to the absolute number, proportion and representativeness of individuals who receive it. Generally, data on such issues are poorly reported in trials and often the data that are reported are not comparable between studies. We extracted data from trials on the proportion of eligible patients who did not take part, these data are presented in Appendix 10. Participation rates were unclear or not reported in 27 studies (28% of the data set). The average participation rate across the remaining studies was 70%, with a range of 13–100%. Interpretation of these data are difficult because of the variation and ambiguity in the exact recruitment procedures employed by each study involved for effective comparison. Sample representativeness was not reported in 39 studies; 50 studies reported study exclusion criteria, including acute and comorbid long-term health conditions.

Effectiveness

Effectiveness is defined as the impact of an intervention on important outcomes, including potential negative effects, QoL and economic outcomes. In this review, the effects of self-care are presented in forest and permutation plots, including any potential detrimental effects on QoL. The validity of the conclusions drawn at each stage remains dependent on the size of the evidence base and its scientific rigour. Limitations in the primary evidence base are considered, where appropriate, and a sensitivity analysis based on evidence quality has been carried out. Limitations in review procedures are discussed in the following chapter (see Chapter 4).

Adoption

The adoption of health behaviour interventions is dependent on the absolute number, proportion and representativeness of the settings and facilitators delivering a programme. Data relating to the proportion and representativeness of the settings used in the primary research studies were rarely reported. We have used subgroup analyses to compare the effects of self-care support delivered in different intervention settings. We extracted detailed information on intervention setting, size and facilitator expertise and present these data in Appendix 10. In our review, the vast majority of interventions (n = 95, 83%) were delivered by qualified health professionals or paraprofessionals (i.e. workers with formal tertiary education or training). Only four interventions were delivered by lay health workers (receiving only informal job-related training). One additional study included a lay health worker as part of a multidisciplinary team.

Implementation

Implementation refers to a study’s fidelity to an intervention protocol. This includes consistency of intervention delivery and the time and costs required to deliver the intervention as intended of the intervention. Twenty-four studies (25%) did not report any process measures. The majority of the remaining studies reported basic data on patient engagement. Lack of data pertaining to facilitator engagement and intervention fidelity means it is difficult to know the extent to which interventions were delivered as intended.

Maintenance

Maintenance in the RE-AIM framework refers to the long-term effects of an intervention on individual patient or organisational outcomes ≥ 6 months after intervention completion. Where multiple data points were reported in the primary studies, we selected outcomes closest to a 12-month follow-up. The mean (SD) follow-up duration for the data extracted for our review was 10.4 (SD 4.6) months.

Self-care support interventions have small but statistically significant effects on children and young people’s quality of life

This review is the first to simultaneously examine the effect of self-care support for LTCs on patient outcomes and health service utilisation in children and young people. Pooled ESs suggest that self-care support has a positive but minimal effect on QoL (ES of 0.17). Evidence is most robust for children and young people with asthma (ES of 0.15) and long-term mental health conditions (ES of 0.17). Lack of evidence for other conditions (or condition clusters) prohibits meaningful assessments of effect.

A pooled effect of 0.17 aligns well with the results of our preceding review, which used identical methods to establish the effects of self-care support for LTCs in adults. 26 In adult populations, self-care support is associated with small but statistically significant effects on QoL (ES of 0.22). 26

Direct comparison of our findings with other reviews of self-care support for children and young people is limited by fundamental differences in remits and scope. A prior review of the clinical effectiveness of self-care support interventions for children and young people with physical health conditions31 reported positive impacts on QoL, but synthesised data narratively and did not present standardised ESs derived from a meta-analysis of intervention effects. The effect of self-care support on the health status of children and young people with mental health conditions has been studied separately; in this instance pooled ESs of 0.20 and 0.12 were reported at 6 and 12 months, respectively. 32

The size and the scope of the evidence base differ between different reviews. These differences are not unusual and reflect both practical and methodological variances. Earlier reviews31,32 adopted different search dates and applied different eligibility criteria, stemming from their need to address different research aims. In line with our protocol, we legitimately excluded studies that failed to report both clinical and economic outcomes. This was because our review was designed to identify those models of self-care support that could reduce health services utilisation and costs, without compromising outcomes for children and young people. Only studies reporting both forms of data could answer this brief. A number of trials of self-care support have reported children and young people’s subjective QoL assessments, but have not simultaneously quantified the impact of these interventions on health service costs. We acknowledge that some evidence with broader relevance to our population may have been excluded by these studies failing to meet our inclusion criteria.

Our up-to-date and comprehensive review makes an important and meaningful contribution to service development and commissioning debates. When QoL was plotted against health service utilisation data, relatively fewer studies reported reductions in both outcomes. This suggests that any intention to use self-care support to reduce utilisation should therefore not raise concerns that these interventions routinely compromise patients’ QoL.

In drawing this conclusion, it is important to remember that study effects are conventionally reported at the level of the group. The available data apply only to those participants consenting to take part in the included research studies. Where reported, study participation rates appeared typical of behavioural intervention trials, but explorations of sample representativeness were limited by inconsistent data and ambiguous reporting. Our review focused solely on children and young people’s QoL and did not consider the impact of self-care support on parental or family outcomes. Thus, it may be prudent for health professionals to monitor the individual impact of self-care support, including any potential effects on these broader contexts, during routine consultations with their patients.

Self-care support interventions lack clear benefits for health service utilisation and cost

Having established that self-care support does not routinely compromise children and young people’s QoL, it becomes necessary to consider its effect on health-care utilisation and costs. On the basis of the current evidence, we cannot reliably conclude that self-care support significantly reduces overall health-care costs.

Analysis of the impact of self-care support on total costs (our primary analysis) was limited by a lack of available data. Pooled ESs suggest a minimal effect (ES of –0.11) on ED visits and a non-significant effect on hospital admissions (ES of –0.06). Lack of a statistically significant effect on hospital admissions endured across different intervention intensities, evidence quality levels and LTCs.

Small reductions in the number of ED visits may well confer multiple advantages on children, young people and their families, not least because accessing emergency care can be associated with acute emotional and logistical stressors. However, at a population level, uncertainty remains regarding the extent to which minimal reductions in emergency use can offset the costs of delivering self-care support and ameliorate the fiscal burdens facing contemporary health-care systems. Our exploratory subgroup analyses suggest that effects may be more pronounced in UK than in non-UK settings and that more intensively facilitated self-care support interventions may be necessary to secure minimal benefits for ED use. In the absence of meta-regression, these subgroup analyses are limited by the fact that many other factors may also differ between study groups.

Any analysis focused on a single aspect of health-care utilisation is vulnerable to error. Single aspects of health-care utilisation represent ‘partial’ cost data, which, by definition, neglect costs and cost changes elsewhere in the system. Arguably, cost shifting may have been a greater risk had substantial reductions in health service utilisation been revealed. In this instance, it would have become crucial to ascertain whether the observed effects reflected genuine reductions in health service use or whether costs had simply been transferred to other health-care sectors or on to patients. With the exception of ED visits and hospital admissions, health service utilisation data were inconsistently reported by the primary studies in our review. This lack of transparent and standardised reporting means that we cannot be certain that other cost changes were not encountered.

Total cost outcome data are necessary to provide policy-makers and service providers with clear evidence of the efficiency (or otherwise) of self-care support. By definition, these data sum costs across all service sectors and include the costs of delivering the intervention that is intended to generate these cost changes. Few primary studies reported total cost data in the current review, prohibiting this more robust analysis.

Health utilisation outcomes rarely distinguish between ‘necessary’ and ‘avoidable’ service use. 26 Such distinctions are not common in the self-care support literature, partly because the difference is conceptually and logistically difficult to assess. 208 An implicit assumption underlying self-care support is that it reduces ‘avoidable’ health service contacts, either by improving a person’s overall health and/or by precipitating more effective crisis responses. Some of the studies in our review distinguished between scheduled and non-scheduled health service use, but did not report care appropriateness per se. We did not distinguish between elective and unplanned admissions in our analysis, nor did we distinguish between legitimate and inappropriate ED use. The effects of self-care support interventions on these different forms of health service utilisation may conceivably be very different.

The optimal assessment of the hypothesis underlying this review would have been to restrict our analysis to the most comprehensive assessment of costs, including those related to NHS service use, social care and other services (e.g. the cost of additional or alternative education for children and young people living with LTCs). Multiple family expenses may be incurred when a child or young person lives with a LTC and arguably the impact of self-care support on families’ out-of-pocket expenses and productivity need to be examined. There are inherent challenges in evaluating the economic effects of self-care support from this broader ‘societal perspective’.

From an operational perspective, it is difficult to foresee how self-care support interventions justified from a societal perspective will be implemented. 209 This is because the budgets for different service sectors are fixed and/or the financial transfers which would be needed between them are not always possible. Although any necessary reallocation of resources can be identified, transfers between sectors are not always considered desirable or feasible. 210 From an evaluative perspective, comprehensive costing of self-care support interventions is rare and generally restricted to formal economic analyses. Our review included 97 studies, of which only 35 reported formal economic analyses. The broader evidence base included in our review is reflective of a larger number of studies that report useful data on health service utilisation. Systematic assessment of this wider literature makes an important and much-needed contribution to policy and service development.

A global economic crisis means that substantial effort continues to be invested in improving the efficiency of health-care systems. Yet, despite self-care being advocated as a key method of increasing service efficiency, there remains uncertainty regarding the scale of the contribution that can be made. Although a previous review has suggested that self-care support interventions may reduce hospital use and total costs in adults, our study has demonstrated potentially smaller effects in children and young people. Understanding the reason for these differences may be an important focus for future research efforts.

Insufficient evidence of an effect of self-care support on children and young people’s health service utilisation raises important questions regarding the primary drivers for its use. Self-care support resonates with multiple policy strategies, including philosophical shifts towards partnership working and the delivery of personalised care. 33 In children and young people, self-care support may offer an early and otherwise unavailable opportunity to negotiate LTC management, maximise adult health outcomes and make a potentially longer-term, sustained contribution to service utilisation. 31,39 Consideration must thus be given to both its processes and outcomes, and the potential breadth of benefits that it will confer. Specific attention should be given to different stakeholder perspectives and whose views – population, policy, professional or patient – are the most important when minimal effects on QoL and health service utilisation are observed.

The impact of self-care support interventions may depend less on intervention intensity and more on its delivery mechanisms

Rigorous evaluation of the efficiency of self-care support interventions for children and young people with LTCs demands concurrent evaluation of patient well-being and health utilisation effects. The suggestion that self-care support may minimally benefit QoL, but not translate into marked benefits for health service use held, across different age groups, intervention intensities and settings. Constraints on the number of data underpinning these results demand some caution in their interpretation.

In line with our protocol and previous review,26 we categorised intervention intensity according to a broad typology and compared pure or facilitated self-care support with more intensively facilitated or case-managed care. The threshold for intensive facilitation took account of both amount of self-care support (> 2 hours or four sessions), as well as the nature of the support provided. This threshold was an arbitrary empirical threshold that provided a reasonable distribution of studies among the different categories.

Reductions in ED use were not consistent across LTCs or intervention type. Preliminary analyses suggest a significant reduction in emergency use for children aged < 13 years, children and young people with asthma and children and young people receiving more intensively facilitated self-care support interventions. However, the existing evidence base is of only moderate size and these different findings will, in part, reflect differences in the number of studies available and the precision of the pooled effects.

Pooled effects suggest a significant benefit for self-care support interventions for asthma that is not confirmed in mental health. Self-care support interventions for children and young people can vary considerably in the extent to which they target different service utilisation behaviours and this potential influence may be meaningful. It is plausible, for example, that, although written action plans to control asthma exacerbations may play a direct role in avoiding ED visits, self-care support for mental health may be more focused on longer-term recovery and patient empowerment. Notably, however, the potential burden of these different intervention models may also differ. Preliminary data in our permutation plots suggest that, although self-care support interventions can reduce utilisation for children and young people with asthma, compromises in their QoL cannot definitively be ruled out. Compromises in QoL were less evident for mental health conditions, although meaningful interpretation is currently limited by a lack of available data.

Our review did not explore differences in the effects of interventions with different content; this information was inconsistently reported by the primary studies in our review. Service developers might usefully explore the process and content of those interventions that did and did not compromise outcomes in the current review to assess the implications of this for future service design. Direct consideration of the aim and purpose of different self-care support interventions, including the rationale for delivering higher-intensity self-care support, may benefit service delivery.

Optimal assessment of the effects of more and less intensive self-care support demands a head-to-head comparison. Only six trials in our review adopted this design. 79,98,99,111,178,191 Meta-regression is possible, but has limited utility in moderate-to-small data sets as a result of a lack of available power. The variability that we observed in intervention descriptions also challenges its use. Lack of standardisation in the terminology and level of detail used to describe self-care support interventions meant that meta-regression had limited function in the context of the current evidence base.

The available evidence suggests that the effect of self-care support on children and young people’s QoL may be optimised by interventions that include the child or young person and deliver at least some of their content to an individual or individual family. Preliminary analyses suggest that face-to-face delivery may be necessary to secure minimal benefits for ED use but, at present, the evidence base does not discriminate between outpatient clinic or community settings. Further research is needed to confirm which approach works best, in what context and for what condition. Without evidence to suggest that health service utilisation is differentially impacted by different delivery models, decisions regarding where or how to deliver self-care may usefully be determined by patient and practitioner preferences and available service resources.

Differences in the magnitude and pattern of effects that were observed across children and young people’s QoL and hospital admissions are subject to multiple interpretations. One explanation is that the magnitude or nature of the improvement in QoL is insufficient to reduce this aspect of children and young people’s utilisation. Another is that children and young people’s subjective health assessments remain somewhat independent of their service use. Self-care in relation to children and young people is known to be complex and conceptually different from that of adult populations. This is, in part, attributable to the role that parents play in managing their child’s condition and the potentially different psychosocial consequences of LTC management for children, their parents and families. 210 Any policy mandates that seek to change children and young people’s health behaviours must consider children’s autonomy in the health-care system and the broader social and family contexts in which decision-making is occurring. Those developing and designing self-care support interventions might usefully consider the extent to which reductions in utilisation are an explicit goal of the intervention, the extent to which health professionals are prepared and willing to transfer responsibility to families211 and the extent to which parents and young people are willing to receive it. 51–53

Our review has identified a potential area of conflict in the delivery of self-care support interventions. Although effects on QoL and ED use may be optimised by delivering interventions to individuals, group-based interventions may be more likely to result in demonstrable reductions to hospital admissions. Group-based models of self-care support have previously been reported to normalise chronic illness, reduce social isolation and develop the social networks of children, young people and their parents,31 while also offering potential cost savings through higher staff-to-child ratios. Any notion that they may also confer benefits on health service utilisation may thus appeal to service providers. However, limitations in the current evidence base mean that this result must be treated with caution and further research is necessary to test this hypothesis prior to significant investment in service development.

The size and scope of the evidence base should be expanded to ascertain the effects of self-care support across a wider range of long-term conditions

Our review identified a much smaller evidence base than our previous review, which used comparable methods to evaluate self-care support interventions for LTCs in adults. The smaller evidence base in this review is consistent with the recognition that the majority of self-care research has been conducted with adult populations. There has been a lack of a cumulative approach to learning from studies of self-care support with children and young people, especially in relation to the health economic and utilisation literature. Prior work has acknowledged a lack of synthesis of the effects of self-care support intervention across different long-term health conditions. 31,32

An important observation is that the majority of self-care support interventions included in our review were designed and delivered to children and young people with asthma. This is perhaps not surprising given its prevalence in the child population. However, the incidence and/or survival rates of other conditions (e.g. diabetes and cystic fibrosis) are also increasing;59–61 and meeting the needs and priorities of these children and young people, their families and NHS services is likely to constitute a crucial element of statutory service planning in the future. The generalisability of our findings to other long-term health conditions is not clear.

Our review identified a moderate evidence base for mental health conditions. This broad condition cluster encompassed a range of potential heterogeneous conditions, including (in relatively equal proportions) depression or anxiety, suicidality or self-harm, psychosis or schizophrenia and eating disorders. The decision to combine data across these different conditions was a methodological one, justified on the basis that this provided a reasonable number of studies for meta-analysis and permitted a broad cumulative assessment of the effects of self-care support interventions for mental health. A finer-grained analysis of the effects of self-care support on the different patient experiences and utilisation pathways that these conditions may precipitate demands the acquisition of a much bigger evidence base.

The generation of new evidence should adopt clear and consistent standards of data reporting, including comprehensive reporting of patient outcomes, utilisation and costs

Our review adopted a comprehensive and rigorous approach to study eligibility judgements and data synthesis. Our ability to conduct some of our analyses has been hampered by poor reporting of outcome data in primary studies. Although our typology of self-care support interventions was relatively simple, its application was complicated by variation in the amount of the detail provided, including lack of transparency regarding intervention personnel and the amount and nature of the support that they provided. More comprehensive, consistent and theory-led reporting of intervention content and processes would facilitate much more effective analyses of specific intervention ingredients.

We identified a notable number of studies (n = 19) that met our review inclusion criteria but failed to provide data amenable to meta-analysis. Deficiencies in outcome reporting are common and are not specific to our review, although the requirement that primary studies reported both QoL and utilisation outcomes meant that the impact of these deficiencies was inevitably more acute. More consistent and comprehensive reporting of data would enable more effective syntheses.

Our primary objective was to assess the ability of self-care support to reduce costs without compromising outcomes for children and young people. This objective does not map neatly onto conventional economic analyses, which focus on incremental cost-effectiveness ratios (ICERs) and associated net mean benefit statistics. Traditionally, interventions that increase costs, but provide significant health benefits for children, might attract support from decision-makers, who would then face decisions about which other interventions (with less attractive cost-effectiveness profiles) might be replaced. The current research aimed to establish whether or not cost savings could be made without comprising patient health.

The primary analysis identified in our protocol was on total costs. We applied liberal inclusion criteria to the cost outcomes included in our forest and permutation plots, and included data where it represented a composite measure of health service costs. However, inconsistencies in data reporting meant that not all studies included all sources of health service or intervention delivery costs. As such, some outcomes may have fallen short of what would conventionally be considered a comprehensive assessment of NHS costs. No clear relationship between patient outcomes and costs was evident in the permutation plot, although only a small number of comparisons was available for meta-analysis and variability across studies was high. Lack of data availability meant that we could not accurately assess the robustness of our secondary analyses, which we based on partial costs.

Our protocol stipulated that our secondary analyses would, where data allowed, sequentially explore the effect of self-care support on inpatient, outpatient, primary care and community care resources. We also intended to assess the effect of self-care support on prescribed medication and patients’ and families’ out-of-pocket expenditures. Lack of consistent measurement and ambiguity in the data available meant that these analyses could not be carried out.

From the patient perspective, any positive effect of self-care support on QoL is likely to be appraised and interpreted in the context of other gains and losses, including the costs incurred in engaging in self-care behaviours. Future studies should thus seek to establish which models of self-care support, if any, are associated with reduced service utilisation and explore, through the collection of comprehensive cost data, potential patterns of cost shifting between services and patients.

Potential differences in the shorter- and longer-term effects of self-care support for children and young people should be considered

Within the limits of the data available, our review suggests that self-care support may have minimal but positive impact on children’s QoL and lack clear benefits for health service utilisation. This finding is based on the short-term follow-up data reported by the primary studies in our review. Where multiple follow-up assessments were conducted, we extracted data closest to a 12-month assessment.

Self-care support for children and young people may have an early and valuable part to play in developing self-efficacy, empowering patients and promoting positive health behaviours for LTC management. Insights into the processes underlying utilisation can be derived from adult studies, which suggest that reductions in health service use may be facilitated by shifting conceptions of reliance on traditional services and translating the acquisition of skills and practices into everyday routines. 26 Prior experiences and methods of contact with services appear to need explicit attention to transition successfully to greater self-care. 212 Giving early legitimacy to personal self-care strategies and modifying perceptions of risk may thus be a key way to reduce later service utilisation. Further research may usefully establish if, and if so, which models of self-care support for children and young people have had longer-term effects on QoL and health utilisation. Modelling the long-term economic consequences of improved health outcomes may be necessary to assess these effects, given the logistical difficulties of prolonged follow-up in clinical trials. Qualitative research into the reasons underpinning sustained health service utilisation in children and young people may also afford valuable insights for future service design.

Our review has treated self-care support as a discrete form of health technology that is bounded and capable of being delivered in a standardised form. The primary studies in our review evaluated interventions that lasted for a limited amount of time (median total duration 5 hours delivered over 12 weeks). An implicit assumption driving many self-care support initiatives is that relatively short-term interventions will lead to permanent behaviour change and deliver cumulative linear benefits for health utilisation as professional support is reduced over time. 26,32 Self-care support in children and young people inevitably occurs in the context of many other transient influences, including critical influences from parents and peers. 51–53,213 Fluctuations and complexity in the transfer of responsibilities from parents to young people are routinely acknowledged in the self-care literature50,214 and introduce a level of uncertainty into the likely durability of intervention effects.

Our review suggested similar effects on QoL and hospital admissions irrespective of the intensity of the self-care support interventions. Intensively facilitated or case-managed self-care support interventions were associated with significant but minimal reductions in ED use. Augmented or ongoing support may thus be a possible method of optimising shorter- and longer-term service benefits, but a critical question then becomes whether or not this additional input is justified on the basis of the additional value it confers.

Clearly, further primary research is indicated to explore whether or not new models of self-care support can achieve more powerful and consistent effects on utilisation. This work should follow standard models for the development of complex interventions215 and draw on relevant behavioural and social science models of patient experience and care access in LTCs. Primary research and meta-synthesis of qualitative data31,32 has identified key elements of self-care support that may be essential for children and young people, including the acquisition of knowledge and skills, child-centred services, peer and professional support and positive beliefs about the effectiveness of their self-care regimes. Synthesising these data with new primary research may facilitate the timely development of new and effective evidence-based services.

Once adequate evidence of impact is achieved, research priorities should transition to the implementation of self-care support at a wider population level. The potential for effective models of self-care support for children and young people to be disseminated widely remains unclear. The studies included in our review typically evaluated the effects of self-care support in small or selected samples. Information pertaining the potential reach and adoption of different intervention models was limited and intervention fidelity inadequately reported. Most of the literature we examined originated from outside the UK. It is not clear how well an international evidence base, accumulated over the last three decades, translates to contemporary NHS settings and cultures. Rigorous research, drawing on implementation science methodologies is required to determine the effectiveness and feasibly of self-care support in the context of routine service provision.

Search strategy

Search strategy

#1 MeSH descriptor: [Self Administration] this term only (653)

#2 self next administer*:ti,ab,kw (2287)

#3 (self next administer* near/2 (questionnaire* or survey* or interview*)):ti,ab,kw (792)

#4 #2 not #3 (1495)

#5 MeSH descriptor: [Self Care] this term only (2833)

#6 (self next care* or selfcaring or selfcare* or selfcaring):ti,ab,kw (4256)

#7 (self next manag* or selfmanag*):ti,ab,kw (2420)

#8 (self next monitor* or selfmonitor* or self report* or selfreport*):ti,ab,kw (18,030)

#9 (self next help* or selfhelp*):ti,ab,kw (1870)

#10 (self next diagnos* or selfdiagnos* or self next assess* or selfassess*):ti,ab,kw (2006)

#11 MeSH descriptor: [Blood Glucose Self-Monitoring] this term only (495)

#12 self next initiat*:ti,ab,kw (79)

#13 MeSH descriptor: [Self Efficacy] this term only (1599)

#14 self next efficacy:ti,ab,kw (3692)

#15 MeSH descriptor: [Self Medication] this term only (81)

#16 (self next medicat* or selfmedicat* or self next remed* or selfremed*):ti,ab,kw (247)

#17 (self next treat* or selftreat* or self next cure* or selfcure* or self next curing or selfcuring):ti,ab,kw (218)

#18 (collaborat* next (care or manag*)):ti,ab,kw (366)

#19 recovery:ti,ab,kw (25,073)

#20 #1 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 (52,340)

#21 MeSH descriptor: [Social Support] this term only (2441)

#22 social next support*:ti,ab,kw (3863)

#23 (group next (support* or advice or advis* or monitor* or intervention* or train* or instruct* or consult* or assist* or educat* or information)):ti,ab,kw (2670)

#24 (peer next (support* or advice or advis* or monitor* or intervention* or train* or instruct* or consult* or assist* or educat* or information)):ti,ab,kw (515)

#25 ((mutual or telephone) next support):ti,ab,kw (294)

#26 (expert next patient* or virtual next communit* or online next communit*):ti,ab,kw (45)

#27 (befriend* or coach* or mentor* or buddy or buddies):ti,ab,kw (1348)

#28 health next trainer*:ti,ab,kw (5)

#29 MeSH descriptor: [Telemedicine] this term only (1013)

#30 (telemedicine or telecare or telenursing or telemonitor* or telehealth or ehealth):ti,ab,kw (1776)

#31 MeSH descriptor: [Remote Consultation] this term only (333)

#32 ((telephon* or remote or phone) near/2 (follow* or support* or consult* or advice or advis* or intervention* or instruct* or assist* or educate or education or information or monitor*)):ti,ab,kw (2831)

#33 MeSH descriptor: [Case Management] this term only (652)

#34 (case next manag* or action next plan* or care next plan* or care next manag* or management next plan* or management next program* or care next program* or goal next setting or individual next goal*):ti,ab,kw (6445)

#35 MeSH descriptor: [Patient Participation] this term only (890)

#36 (involv* or participat* or collaborat*):ti,ab,kw (72,418)

#37 MeSH descriptor: [Patient Education as Topic] this term only (6657)

#38 patient next education:ti,ab,kw (8086)

#39 (patient near/2 (educat* or advice or advis* or instruct* or train* or coach*)):ti,ab,kw (8784)

#40 ((consumer next health or patient) next information):ti,ab,kw (553)

#41 (nurse near/2 educator*):ti,ab,kw (73)

#42 ((financial or monetary or money) near/2 (incentive* or competition* or contest* or lotter* or reward* or prize*)):ti,ab,kw (513)

#43 (contingent next payment* or deposit next contract*):ti,ab,kw (17)

#44 MeSH descriptor: [Decision Making] this term only (1692)

#45 (decision* near/2 (shared or support* or aid or aids or making)):ti,ab,kw (7150)

#46 (goal next set* or individual next goal*):ti,ab,kw (446)

#47 MeSH descriptor: [Diet] explode all trees (12,385)

#48 MeSH descriptor: [Exercise] explode all trees (14,181)

#49 MeSH descriptor: [Motivational Interviewing] this term only (136)

#50 MeSH descriptor: [Cognitive Therapy] this term only (5146)

#51 (CBT or cognitive next therap* or cognitive next behav*):ti,ab,kw (9693)

#52 MeSH descriptor: [Teaching Materials] this term only (383)

#53 MeSH descriptor: [Pamphlets] this term only (623)

#54 (educational next material* or leaflet* or booklet* or toolkit*):ti,ab,kw (1977)

#55 MeSH descriptor: [Bibliotherapy] this term only (105)

#56 MeSH descriptor: [Internet] explode all trees (1900)

#57 MeSH descriptor: [Parenting] this term only (669)

#58 parent* next training:ti,ab,kw (382)

#59 MeSH descriptor: [Behavior Therapy] this term only (3633)

#60 ((behavior* or behaviour*) next (manag* or modif*)):ti,ab,kw (950)

#61 MeSH descriptor: [Psychotherapy, Brief] this term only (708)

#62 MeSH descriptor: [Psychotherapy, Group] this term only (1533)

#63 psychoeducat*:ti,ab,kw (1071)

#64 MeSH descriptor: [Problem Solving] this term only (1209)

#65 (skills next training or coping next skill* or empower*):ti,ab,kw (2354)

#66 MeSH descriptor: [Adaptation, Psychological] this term only (3112)

#67 MeSH descriptor: [Communication] explode all trees (9893)

#68 MeSH descriptor: [Mindfulness] this term only (32)

#69 MeSH descriptor: [Patient Access to Records] this term only (17)

#70 MeSH descriptor: [Cell Phones] this term only (323)

#71 MeSH descriptor: [Telephone] this term only (1409)

#72 MeSH descriptor: [Text Messaging] this term only (144)

#73 MeSH descriptor: [Reminder Systems] this term only (618)

#74 text next messag*:ti,ab,kw (427)

#75 {or #21-#74} (136,490)

#76 MeSH descriptor: [Child] explode all trees (151)

#77 MeSH descriptor: [Adolescent] this term only (77,091)

#78 (child or children or schoolchild* or baby or babies or infant or infants or toddler* or teen* or teenager* or adolescen* or young next person* or young next people or youth or youngster* or juvenile* or paediatric or pediatric):ti,ab,kw (163,314)

#79 MeSH descriptor: [Parents] explode all trees (2723)

#80 MeSH descriptor: [Caregivers] this term only (1265)

#81 (parent* or mother* or father* or caregiver* or carer* or guardian* or advocate* or family):ti,ab,kw (40,218)

#82 #76 or #77 or #78 or #79 or #80 or #81 (182,717)

#83 MeSH descriptor: [Diabetes Mellitus] explode all trees (16,726)

#84 MeSH descriptor: [Cystic Fibrosis] this term only (1102)

#85 mucoviscidosis:ti,ab,kw (33)

#86 MeSH descriptor: [Asthma] explode all trees (9404)

#87 MeSH descriptor: [Attention Deficit and Disruptive Behavior Disorders] explode all trees (1865)

#88 MeSH descriptor: [Hyperkinesis] this term only (168)

#89 MeSH descriptor: [Anxiety] explode all trees (5226)

#90 MeSH descriptor: [Depression] this term only (5541)

#91 MeSH descriptor: [Depressive Disorder] this term only (4754)

#92 MeSH descriptor: [Self Mutilation] this term only (25)

#93 MeSH descriptor: [Self-Injurious Behavior] this term only (205)

#94 MeSH descriptor: [Epilepsy] explode all trees (2311)

#95 MeSH descriptor: [Conduct Disorder] explode all trees (179)

#96 MeSH descriptor: [Mood Disorders] explode all trees (9310)

#97 MeSH descriptor: [Obsessive-Compulsive Disorder] explode all trees (662)

#98 MeSH descriptor: [Stress Disorders, Post-Traumatic] this term only (972)

#99 MeSH descriptor: [Psychotic Disorders] explode all trees (1562)

#100 MeSH descriptor: [Eating Disorders] explode all trees (838)

#101 MeSH descriptor: [Schizophrenia] explode all trees (4966)

#102 MeSH descriptor: [Bipolar Disorder] explode all trees (1601)

#103 MeSH descriptor: [Phobic Disorders] this term only (847)

#104 MeSH descriptor: [Panic Disorder] this term only (758)

#105 school next refusal:ti,ab,kw (14)

#106 MeSH descriptor: [Mental Health] this term only (689)

#107 {or #83-#106} (57,988)

#108 MeSH descriptor: [Economics] this term only (58)

#109 MeSH descriptor: [Costs and Cost Analysis] explode all trees (23,270)

#110 MeSH descriptor: [Value of Life] this term only (144)

#111 MeSH descriptor: [Economics, Hospital] explode all trees (1655)

#112 MeSH descriptor: [Economics, Medical] this term only (38)

#113 MeSH descriptor: [Economics, Nursing] this term only (17)

#114 MeSH descriptor: [Economics, Pharmaceutical] this term only (236)

#115 MeSH descriptor: [Economics, Dental] this term only (3)

#116 (econom* or cost or costs or costly or costing or price or prices or pricing or pharmacoeconomic*):ti,ab,kw (48,142)

#117 (expenditure* not energy):ti,ab (762)

#118 (value next money):ti,ab,kw (2)

#119 budget*:ti,ab,kw (418)

#120 MeSH descriptor: [Hospitalization] this term only (4574)

#121 MeSH descriptor: [Length of Stay] this term only (6734)

#122 MeSH descriptor: [Patient Admission] this term only (578)

#123 MeSH descriptor: [Patient Readmission] this term only (811)

#124 MeSH descriptor: [Health Resources] this term only (513)

#125 (length near/2 stay):ti,ab,kw (10,176)

#126 (duration near/2 stay):ti,ab,kw (372)

#127 (hospital next (access* or uptake or visit* or attendance* or admission* or admit* or episode*)):ti,ab,kw (3243)

#128 hospital next cost*:ti,ab,kw (2548)

#129 (time near/2 discharg*):ti,ab,kw (896)

#130 hospital next day*:ti,ab,kw (508)

#131 ((patient* or inpatient* or in-patient*) next (cost* or stay*)):ti,ab,kw (772)

#132 (number near/2 (nights or days)):ti,ab,kw (1889)

#133 (primary next care next (visit* or contact* or attendance* or admission* or episode*)):ti,ab,kw (139)

#134 (surgery next (visit* or contact* or attendance* or admission* or episode*)):ti,ab,kw (36)

#135 (GP next (access or uptake or visit* or contact* or attendance* or admission* or episode*)):ti,ab,kw (65)

#136 ((uptake or access) next (service* or care or intervention*)):ti,ab,kw (39)

#137 ((clinic or surgery or hospital or “accident and emergency”) near/2 (work-flow or work next flow)):ti,ab,kw (2)

#138 (consultation* near/2 (time or length)):ti,ab,kw (116)

#139 (hospitalisation* or hospitalization* or rehospitalisation* or rehospitalization* or re-hospitalisation* or re-hospitalization*):ti,ab,kw (15,517)

#140 (hrql or hrqol or “h qol” or h-qol or hql or hqol):ti,ab,kw (1887)

#141 MeSH descriptor: [Quality-Adjusted Life Years] this term only (3773)

#142 (qaly* or “quality adjusted life” or “quality of life” or “life quality”):ti,ab,kw (37,642)

#143 {or #108-#142} (97,676)

#144 #20 or #75 (169,797)

#145 #144 and #82 and #107 and #143 (1739)

#146 “Cool Kids”:ti,ab,kw (3)

#147 “Sweet talk”:ti,ab,kw (2)

#148 “Timid to Tiger”:ti,ab,kw (0)

#149 “problem solving for life”:ti,ab,kw (2)

#150 “Incredible Years”:ti,ab,kw (67)

#151 “Triple P”:ti,ab,kw (77)

#152 friends next program*:ti,ab,kw (12)

#153 #146 or #147 or #148 or #149 or #150 or #151 or #152 (162)

#154 #145 or #153 (1894)

Search strategy

1. self administer*.ti,ab. (30,354)

2. (self administer$ adj2 (questionnaire$ or survey$ or interview$)).ti,ab. (18,534)

3. 1 not 2 (11,820)

4. self care/ (33,623)

5. (self care$ or selfcaring or selfcare$ or selfcaring).ti,ab. (14,723)

6. (self manag$ or selfmanag$).ti,ab. (13,745)

7. (self monitor$ or selfmonitor$ or self report$ or selfreport$).ti,ab. (120,139)

8. (self help$ or selfhelp$).ti,ab. (6234)

9. (self diagnos$ or selfdiagnos$ or self assess$ or selfassess$).ti,ab. (14,481)

10. blood glucose self-monitoring/ (9642)

11. self initiat$.ti,ab. (1171)

12. self concept/ (65,378)

13. self efficacy.ti,ab. (17,042)

14. Drug self administration/ (8067)

15. (self medicat$ or selfmedicat$ or self remed$ or selfremed$).ti,ab. (4434)

16. (self treat$ or selftreat$ or self cure$ or selfcure$ or self curing or selfcuring).ti,ab. (2748)

17. (collaborat$ adj (care or manag$)).ti,ab. (1749)

18. recovery.ti,ab. (399,435)

19. 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 (666,873)

20. social support/ (60,564)

21. social support$.ti,ab. (29,089)

22. (group adj (support$ or advice or advis$ or monitor$ or intervention$ or train$ or instruct$ or consult$ or assist$ or educat$ or information)).ti,ab. (7925)

23. (peer adj (support$ or advice or advis$ or monitor$ or intervention$ or train$ or instruct$ or consult$ or assist$ or educat$ or information)).ti,ab. (4169)

24. ((mutual or telephone) adj support).ti,ab. (1115)

25. (expert patient$ or virtual communit$ or online communit$).ti,ab. (914)

26. (befriend$ or coach$ or mentor$ or buddy or buddies).ti,ab. (20,445)

27. health trainer$.ti,ab. (65)

28. telemedicine/ (12,152)

29. (telemedicine or telecare or telenursing or telemonitor$ or telehealth or ehealth).ti,ab. (11,512)

30. teleconsultation/ (6439)

31. ((telephon$ or remote or phone) adj2 (follow$ or support$ or consult$ or advice or advis$ or intervention$ or instruct$ or assist$ or educate or education or information or monitor$)).ti,ab. (13,807)

32. case management/ (8130)

33. (case manag$ or action plan$ or care plan$ or care manag$ or management plan$ or management program$ or care program$ or goal setting or individual goal$).ti,ab. (66,439)

34. Patient Participation/ (17,929)

35. (involv$ or participat$ or collaborat$).ti,ab. (2,443,918)

36. patient education/ (88,903)

37. patient education.ti,ab. (16,471)

38. (patient adj2 (educat$ or advice or advis$ or instruct$ or instruct$ or train$ or coach$)).ti,ab. (24,060)

39. ((consumer health or patient) adj information).ti,ab. (7877)

40. (nurse adj2 educator$).ti,ab. (2777)

41. ((financial or monetary or money) adj2 (incentive$ or competition$ or contest$ or lotter$ or reward$ or prize$)).ti,ab. (5470)

42. (contingent payment$ or deposit contract$).ti,ab. (33)

43. decision making/ (149,178)

44. (decision$ adj2 (shared or support$ or aid or aids or making)).ti,ab. (112,354)

45. (goal set$ or individual goal$).ti,ab. (3176)

46. exp Diet/ or exp exercise/ (433,362)

47. Motivational Interviewing/ (1362)

48. Cognitive therapy/ (35,839)

49. (CBT or cognitive therap$ or cognitive behav$).ti,ab. (28,386)

50. Teaching/ or Publication/ (194,835)

51. (educational material$ or leaflet$ or booklet$ or toolkit$).ti,ab. (31,734)

52. Bibliotherapy/ (68)

53. Internet/ or Social Media/ (80,342)

54. exp Child Parent Relations/ (68,127)

55. (parenting or parent$ training).ti,ab. (13,012)

56. exp Behavior Therapy/ (37,925)

57. ((behavior$ or behaviour$) adj (manag$ or modif$)).ti,ab. (6548)

58. exp Psychotherapy/ (195,095)

59. Group therapy/ (18,204)

60. psychoeducat$.ti,ab. (4261)

61. Problem Solving/ (27,244)

62. (skills training or coping skill$ or empower$).ti,ab. (25,022)

63. Adaptative Behavior/ (0)

64. exp interpersonal communication/ (417,154)

65. Mindfulness/ (1015)

66. exp Patient Right/ (91,293)

67. mobile phone/ or telephone/ or Text Messaging/ or Reminder System/ (37,748)

68. text messag$.ti,ab. (1632)

69. or/20-68 (4,059,777)

70. exp Child/ (2,160,034)

71. exp juvenile/ (2,820,035)

72. (child or children or schoolchild$ or baby or babies or infant or infants or toddler$ or teen$ or teenager$ or adolescen$ or young person$ or young people or youth or youngster$ or juvenile$ or paediatric or pediatric).ti,ab. (1,761,362)

73. exp Parent/ (168,626)

74. Caregiver/ (43,942)

75. (parent$ or mother$ or father$ or caregiver$ or carer$ or guardian$ or advocate$ or family).ti,ab. (1,237,658)

76. 70 or 71 or 72 or 73 or 74 or 75 (4,140,594)

77. exp Diabetes Mellitus/ (642,426)

78. Cystic Fibrosis/ (50,945)

79. mucoviscidosis.mp. (1548)

80. exp Asthma/ (196,806)

81. Attention Deficit Disorder/ (39,064)

82. Hyperkinesia/ (4305)

83. exp Anxiety/ (129,925)

84. Depression/ (260,270)

85. Automutilation/ (11,299)

86. exp Epilepsy/ (181,678)

87. exp Conduct Disorder/ (4717)

88. exp Mood Disorders/ (361,339)

89. exp Obsessive-Compulsive Disorder/ (29,047)

90. Posttraumatic Stress Disorder/ (37,336)

91. exp Psychosis/ (227,323)

92. exp Eating Disorder/ (37,814)

93. exp Schizophrenia/ (148,365)

94. exp Bipolar Disorder/ (42,348)

95. exp Phobia/ (22,859)

96. Panic/ (18,214)

97. school refusal.ti,ab. (310)

98. exp Mental Health/ (90,904)

99. 77 or 78 or 79 or 80 or 81 or 82 or 83 or 84 or 85 or 86 or 87 or 88 or 89 or 90 or 91 or 92 or 93 or 94 or 95 or 96 or 97 or 98 (1,802,468)

100. exp economics/ (222,944)

101. exp Cost/ (265,768)

102. Health care planning/ (80,549)

103. (econom$ or cost or costs or costly or costing or price or prices or pricing or pharmacoeconomic$).ti,ab. (653,899)

104. (expenditure$ not energy).tw. (25,636)

105. (value adj money).tw. (6)

106. budget$.tw. (25,819)

107. hospitalization/ or “length of stay”/ or patient admission/ or patient readmission/ (401,452)

108. ((healthcare or health) adj resource$).ti,ab. (8383)

109. (length adj2 stay).ti,ab. (50,442)

110. (duration adj2 stay).ti,ab. (3162)

111. (hospital adj (access$ or uptake or visit$ or attendance$ or admission$ or admit$ or episode$)).ti,ab. (39,882)

112. hospital cost$.ti,ab. (7701)

113. (time adj2 discharg$).ti,ab. (6339)

114. hospital day$.ti,ab. (6826)

115. ((patient$ or inpatient$ or in-patient$) adj (cost$ or stay$)).ti,ab. (8915)

116. (number adj2 (nights or days)).ti,ab. (12,521)

117. (primary care adj (visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (1236)

118. (surgery adj (visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (342)

119. (GP adj (access or uptake or visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (614)

120. ((uptake or access) adj (service$ or care or intervention$)).ti,ab. (1011)

121. ((clinic or surgery or hospital or “accident and emergency”) adj2 (work-flow or work flow)).ti,ab. (13)

122. (consultation$ adj2 (time or length)).ti,ab. (1466)

123. (hospitalisation$ or hospitalization$ or rehospitalisation$ or rehospitalization$ or re-hospitalisation$ or re-hospitalization$).ti,ab. (152,093)

124. (hrql or hrqol or h qol or hql or hqol).ti,ab. (16,145)

125. quality adjusted life year/ (13,443)

126. (qaly$ or quality adjusted life or quality of life or life quality).tw. (242,535)

127. 100 or 101 or 102 or 103 or 104 or 105 or 106 or 107 or 108 or 109 or 110 or 111 or 112 or 113 or 114 or 115 or 116 or 117 or 118 or 119 or 120 or 121 or 122 or 123 or 124 or 125 or 126 (1,622,052)

128. 19 or 69 (4,503,231)

129. 128 and 76 and 99 and 127 (21,601)

130. Cool Kids.ti,ab. (10)

131. Sweet talk.ti,ab. (18)

132. Timid to Tiger.ti,ab. (0)

133. “problem solving for life”.ti,ab. (11)

134. Incredible Years.ti,ab. (109)

135. Triple P.ti,ab. (171)

136. friends program$.ti,ab. (17)

137. 130 or 131 or 132 or 133 or 134 or 135 or 136 (327)

138. 129 or 137 (21,901)

139. (editorial or comment or letter).pt. (1,340,488)

140. 138 not 139 (21,728)

141. limit 140 to embase (17,780)

Search strategy

1. self administration/ (9309)

2. self administer*.ti,ab. (24,903)

3. (self administer$ adj2 (questionnaire$ or survey$ or interview$)).ti,ab. (15,609)

4. 2 not 3 (9294)

5. self care/ (24,322)

6. (self care$ or selfcaring or selfcare$ or selfcaring).ti,ab. (10,990)

7. (self manag$ or selfmanag$).ti,ab. (9689)

8. (self monitor$ or selfmonitor$ or self report$ or selfreport$).ti,ab. (95,133)

9. (self help$ or selfhelp$).ti,ab. (4821)

10. (self diagnos$ or selfdiagnos$ or self assess$ or selfassess$).ti,ab. (10,513)

11. blood glucose self-monitoring/ (4401)

12. self initiat$.ti,ab. (955)

13. self efficacy/ (12,604)

14. self efficacy.ti,ab. (14,136)

15. self medication/ (4026)

16. (self medicat$ or selfmedicat$ or self remed$ or selfremed$).ti,ab. (2938)

17. (self treat$ or selftreat$ or self cure$ or selfcure$ or self curing or selfcuring).ti,ab. (2268)

18. (collaborat$ adj (care or manag$)).ti,ab. (1381)

19. recovery.ti,ab. (315,800)

20. 1 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 (495,898)

21. social support/ (52,876)

22. social support$.ti,ab. (23,715)

23. (group adj (support$ or advice or advis$ or monitor$ or intervention$ or train$ or instruct$ or consult$ or assist$ or educat$ or information)).ti,ab. (5138)

24. (peer adj (support$ or advice or advis$ or monitor$ or intervention$ or train$ or instruct$ or consult$ or assist$ or educat$ or information)).ti,ab. (3223)

25. ((mutual or telephone) adj support).ti,ab. (809)

26. (expert patient$ or virtual communit$ or online communit$).ti,ab. (662)

27. (befriend$ or coach$ or mentor$ or buddy or buddies).ti,ab. (16,424)

28. health trainer$.ti,ab. (45)

29. telemedicine/ (11,775)

30. (telemedicine or telecare or telenursing or telemonitor$ or telehealth or ehealth).ti,ab. (9452)

31. remote consultation/ (3667)

32. ((telephon$ or remote or phone) adj2 (follow$ or support$ or consult$ or advice or advis$ or intervention$ or instruct$ or assist$ or educate or education or information or monitor$)).ti,ab. (9595)

33. case management/ (8506)

34. (case manag$ or action plan$ or care plan$ or care manag$ or management plan$ or management program$ or care program$ or goal setting or individual goal$).ti,ab. (50,370)

35. Patient Participation/ (18,448)

36. (involv$ or participat$ or collaborat$).ti,ab. (2,006,956)

37. patient education as topic/ (71,385)

38. patient education.ti,ab. (12,012)

39. (patient adj2 (educat$ or advice or advis$ or instruct$ or instruct$ or train$ or coach$)).ti,ab. (17,109)

40. ((consumer health or patient) adj information).ti,ab. (5353)

41. (nurse adj2 educator$).ti,ab. (2644)

42. ((financial or monetary or money) adj2 (incentive$ or competition$ or contest$ or lotter$ or reward$ or prize$)).ti,ab. (4578)

43. (contingent payment$ or deposit contract$).ti,ab. (25)

44. decision making/ (70,170)

45. (decision$ adj2 (shared or support$ or aid or aids or making)).ti,ab. (87,782)

46. (goal set$ or individual goal$).ti,ab. (2251)

47. exp Diet/ or exp exercise/ (313,869)

48. Motivational Interviewing/ (393)

49. Cognitive therapy/ (16,603)

50. (CBT or cognitive therap$ or cognitive behav$).ti,ab. (19,378)

51. Teaching Materials/ or Pamphlets/ (8753)

52. (educational material$ or leaflet$ or booklet$ or toolkit$).ti,ab. (24,101)

53. Bibliotherapy/ (342)

54. Internet/ or Blogging/ or Social Media/ (52,994)

55. Parenting/ (10,836)

56. parent$ training.ti,ab. (778)

57. Behavior Therapy/ (23,664)

58. ((behavior$ or behaviour$) adj (manag$ or modif$)).ti,ab. (5083)

59. Psychotherapy, Brief/ (2781)

60. Psychotherapy, Group/ (11,975)

61. psychoeducat$.ti,ab. (2827)

62. Problem Solving/ (21,273)

63. (skills training or coping skill$ or empower$).ti,ab. (19,438)

64. Adaptation, Psychological/ (76,111)

65. exp Communication/ (374,050)

66. Mindfulness/ (345)

67. Patient Access to Records/ (871)

68. Cell Phones/ or Telephones/ or Text Messaging/ or Reminder Systems/ (16,546)

69. text messag$.ti,ab. (1261)

70. or/21-69 (3,069,577)

71. exp Child/ (1,568,167)

72. Adolescent/ (1,639,965)

73. (child or children or schoolchild$ or baby or babies or infant or infants or toddler$ or teen$ or teenager$ or adolescen$ or young person$ or young people or youth or youngster$ or juvenile$ or paediatric or pediatric).ti,ab. (1,420,691)

74. exp Parents/ (74,561)

75. Caregivers/ (22,451)

76. (parent$ or mother$ or father$ or caregiver$ or carer$ or guardian$ or advocate$ or family).ti,ab. (1,021,910)

77. 71 or 72 or 73 or 74 or 75 or 76 (3,693,299)

78. exp Diabetes Mellitus/ (322,132)

79. Cystic Fibrosis/ (28,835)

80. mucoviscidosis.mp. (1399)

81. exp Asthma/ (107,539)

82. exp “Attention Deficit and Disruptive Behavior Disorders”/ (23,586)

83. Hyperkinesis/ (3695)

84. exp Anxiety/ (57,586)

85. Depression/ (79,319)

86. Depressive Disorder/ (59,030)

87. Self Mutilation/ or Self-Injurious Behavior/ (8004)

88. exp Epilepsy/ (131,823)

89. exp Conduct Disorder/ (2286)

90. exp Mood Disorders/ (117,649)

91. exp Obsessive-Compulsive Disorder/ (11,558)

92. Stress Disorders, Post-Traumatic/ (21,973)

93. exp Psychotic Disorders/ (39,064)

94. exp Eating Disorders/ (22,708)

95. exp Schizophrenia/ (86,432)

96. exp Bipolar Disorder/ (32,171)

97. Phobic Disorders/ (9386)

98. Panic Disorder/ (6032)

99. school refusal.ti,ab. (207)

100. Mental Health/ (22,673)

101. 78 or 79 or 80 or 81 or 82 or 83 or 84 or 85 or 86 or 87 or 88 or 89 or 90 or 91 or 92 or 93 or 94 or 95 or 96 or 97 or 98 or 99 or 100 (1,006,004)

102. economics/ (26,582)

103. exp “Costs and Cost Analysis”/ (185,512)

104. value of life/ (5422)

105. exp economics, hospital/ or economics, medical/ or economics, nursing/ or economics, pharmaceutical/ or economics, dental/ (36,516)

106. (econom$ or cost or costs or costly or costing or price or prices or pricing or pharmacoeconomic$).ti,ab. (507,710)

107. (expenditure$ not energy).tw. (19,526)

108. (value adj money).tw. (5)

109. budget$.tw. (19,998)

110. hospitalization/ or “length of stay”/ or patient admission/ or patient readmission/ (148,295)

111. health resources/ (9102)

112. (length adj2 stay).ti,ab. (30,699)

113. (duration adj2 stay).ti,ab. (2182)

114. (hospital adj (access$ or uptake or visit$ or attendance$ or admission$ or admit$ or episode$)).ti,ab. (27,387)

115. hospital cost$.ti,ab. (5555)

116. (time adj2 discharg$).ti,ab. (4083)

117. hospital day$.ti,ab. (4364)

118. ((patient$ or inpatient$ or in-patient$) adj (cost$ or stay$)).ti,ab. (5452)

119. (number adj2 (nights or days)).ti,ab. (8981)

120. (primary care adj (visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (953)

121. (surgery adj (visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (232)

122. (GP adj (access or uptake or visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (407)

123. ((uptake or access) adj (service$ or care or intervention$)).ti,ab. (721)

124. ((clinic or surgery or hospital or “accident and emergency”) adj2 (work-flow or work flow)).ti,ab. (6)

125. (consultation$ adj2 (time or length)).ti,ab. (1013)

126. (hospitalisation$ or hospitalization$ or rehospitalisation$ or rehospitalization$ or re-hospitalisation$ or re-hospitalization$).ti,ab. (101,245)

127. (hrql or hrqol or h qol or hql or hqol).ti,ab. (10,897)

128. quality adjusted life year/ (7363)

129. (qaly$ or quality adjusted life or quality of life or life quality).tw. (163,414)

130. 102 or 103 or 104 or 105 or 106 or 107 or 108 or 109 or 110 or 111 or 112 or 113 or 114 or 115 or 116 or 117 or 118 or 119 or 120 or 121 or 122 or 123 or 124 or 125 or 126 or 127 or 128 or 129 (993,754)

131. 20 or 70 (3,436,791)

132. 131 and 77 and 101 and 130 (10,689)

133. Cool Kids.ti,ab. (7)

134. Sweet talk.ti,ab. (15)

135. Timid to Tiger.ti,ab. (1)

136. “problem solving for life”.ti,ab. (9)

137. Incredible Years.ti,ab. (95)

138. Triple P.ti,ab. (136)

139. friends program$.ti,ab. (11)

140. 133 or 134 or 135 or 136 or 137 or 138 or 139 (266)

141. 132 or 140 (10,943)

142. animals/ not humans/ (3,907,576)

143. 141 not 142 (10,938)

144. (editorial or comment or letter).pt. (1,386,778)

145. 143 not 144 (10,870)

Search strategy

1. Drug self administration/ (1547)

2. self administer*.ti,ab. (10,130)

3. (self administer$ adj2 (questionnaire$ or survey$ or interview$)).ti,ab. (5254)

4. 2 not 3 (4876)

5. self care skills/ (3532)

6. (self care$ or selfcaring or selfcare$ or selfcaring).ti,ab. (6125)

7. (self manag$ or selfmanag$).ti,ab. (6074)

8. (self monitor$ or selfmonitor$ or self report$ or selfreport$).ti,ab. (87,437)

9. (self help$ or selfhelp$).ti,ab. (6904)

10. (self diagnos$ or selfdiagnos$ or self assess$ or selfassess$).ti,ab. (6241)

11. blood glucose self-monitoring/ (0)

12. self initiat$.ti,ab. (1179)

13. self efficacy/ (15,881)

14. self efficacy.ti,ab. (25,391)

15. self medication/ (562)

16. (self medicat$ or selfmedicat$ or self remed$ or selfremed$).ti,ab. (1279)

17. (self treat$ or selftreat$ or self cure$ or selfcure$ or self curing or selfcuring).ti,ab. (383)

18. (collaborat$ adj (care or manag$)).ti,ab. (835)

19. recovery.ti,ab. (46,680)

20. 1 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 (189,281)

21. social support/ (28,230)

22. social support$.ti,ab. (34,367)

23. (group adj (support$ or advice or advis$ or monitor$ or intervention$ or train$ or instruct$ or consult$ or assist$ or educat$ or information)).ti,ab. (6330)

24. (peer adj (support$ or advice or advis$ or monitor$ or intervention$ or train$ or instruct$ or consult$ or assist$ or educat$ or information)).ti,ab. (3732)

25. ((mutual or telephone) adj support).ti,ab. (868)

26. (expert patient$ or virtual communit$ or online communit$).ti,ab. (1488)

27. (befriend$ or coach$ or mentor$ or buddy or buddies).ti,ab. (22,379)

28. health trainer$.ti,ab. (17)

29. telemedicine/ (2763)

30. (telemedicine or telecare or telenursing or telemonitor$ or telehealth or ehealth).ti,ab. (1937)

31. (remote consult$ or teleconsult$).ti,ab. (105)

32. ((telephon$ or remote or phone) adj2 (follow$ or support$ or consult$ or advice or advis$ or intervention$ or instruct$ or assist$ or educate or education or information or monitor$)).ti,ab. (3067)

33. exp case management/ (2944)

34. (case manag$ or action plan$ or care plan$ or care manag$ or management plan$ or management program$ or care program$ or goal setting or individual goal$).ti,ab. (22,094)

35. Client participation/ (1459)

36. (involv$ or participat$ or collaborat$).ti,ab. (550,207)

37. client education/ (3154)

38. patient education.ti,ab. (2186)

39. (patient adj2 (educat$ or advice or advis$ or instruct$ or instruct$ or train$ or coach$)).ti,ab. (3416)

40. ((consumer health or patient) adj information).ti,ab. (745)

41. (nurse adj2 educator$).ti,ab. (684)

42. ((financial or monetary or money) adj2 (incentive$ or competition$ or contest$ or lotter$ or reward$ or prize$)).ti,ab. (3247)

43. (contingent payment$ or deposit contract$).ti,ab. (31)

44. exp decision making/ (70,405)

45. (decision$ adj2 (shared or support$ or aid or aids or making)).ti,ab. (64,722)

46. (goal set$ or individual goal$).ti,ab. (4646)

47. exp Diets/ or exp exercise/ (26,852)

48. Motivational Interviewing/ (1454)

49. Cognitive therapy/ (11,872)

50. (CBT or cognitive therap$ or cognitive behav$).ti,ab. (34,514)

51. exp Instructional Materials/ (0)

52. (educational material$ or leaflet$ or booklet$ or toolkit$).ti,ab. (4797)

53. Bibliotherapy/ (610)

54. Internet/ or Social Media/ (24,179)

55. exp Parenting/ (76,945)

56. parent$ training.ti,ab. (2578)

57. exp Behavior Therapy/ (17,292)

58. ((behavior$ or behaviour$) adj (manag$ or modif$)).ti,ab. (7480)

59. Psychotherapy, Brief/ (0)

60. exp Group therapy/ (20,117)

61. psychoeducat$.ti,ab. (6489)

62. exp Problem Solving/ (28,843)

63. (skills training or coping skill$ or empower$).ti,ab. (28,916)

64. Adaptative Behavior/ (0)

65. exp Communication/ (191,116)

66. Mindfulness/ (4018)

67. Client Records/ (677)

68. Cellular Phones/ or Telephone Systems/ or Messages/ (9003)

69. text messag$.ti,ab. (940)

70. or/21-69 (1,048,326)

71. childhood birth 12 yrs.ag. (431,936)

72. adolescence 13 17 yrs.ag. (340,365)

73. (child or children or schoolchild$ or baby or babies or infant or infants or toddler$ or teen$ or teenager$ or adolescen$ or young person$ or young people or youth or youngster$ or juvenile$ or paediatric or pediatric).ti,ab. (686,296)

74. exp Parents/ (74,045)

75. Caregivers/ (20,108)

76. (parent$ or mother$ or father$ or caregiver$ or carer$ or guardian$ or advocate$ or family).ti,ab. (474,588)

77. 71 or 72 or 73 or 74 or 75 or 76 (1,087,542)

78. Diabetes Mellitus/ (4020)

79. Cystic Fibrosis/ (687)

80. mucoviscidosis.mp. (4)

81. Asthma/ (3816)

82. exp Attention Deficit Disorder/ (19,781)

83. Hyperkinesis/ (7565)

84. exp Anxiety/ (53,810)

85. Depression/ (22,138)

86. exp Major Depression/ (97,934)

87. Self-Injurious Behavior/ (2077)

88. exp Epilepsy/ (20,442)

89. Conduct Disorder/ (3575)

90. exp Affective Disorders/ (125,594)

91. Obsessive Compulsive Disorder/ (10,627)

92. Posttraumatic Stress Disorder/ (23,065)

93. exp Psychosis/ (95,067)

94. exp Eating Disorder/ (24,088)

95. exp Schizophrenia/ (74,719)

96. exp Bipolar Disorder/ (20,574)

97. exp Phobias/ (11,206)

98. Panic Disorder/ (6929)

99. school refusal.ti,ab. (478)

100. exp Mental Health/ (44,641)

101. 78 or 79 or 80 or 81 or 82 or 83 or 84 or 85 or 86 or 87 or 88 or 89 or 90 or 91 or 92 or 93 or 94 or 95 or 96 or 97 or 98 or 99 or 100 (422,609)

102. exp economics/ (18,357)

103. exp costs/ (19,890)

104. (econom$ or cost or costs or costly or costing or price or prices or pricing or pharmacoeconomic$).ti,ab. (163,542)

105. (expenditure$ not energy).tw. (5744)

106. (value adj money).tw. (11)

107. budget$.tw. (6378)

108. hospitalization/ or “length of stay”/ or patient admission/ or patient readmission/ (8465)

109. ((healthcare or health) adj resource$).ti,ab. (1564)

110. (length adj2 stay).ti,ab. (3614)

111. (duration adj2 stay).ti,ab. (240)

112. (hospital adj (access$ or uptake or visit$ or attendance$ or admission$ or admit$ or episode$)).ti,ab. (3798)

113. hospital cost$.ti,ab. (252)

114. (time adj2 discharg$).ti,ab. (578)

115. hospital day$.ti,ab. (388)

116. ((patient$ or inpatient$ or in-patient$) adj (cost$ or stay$)).ti,ab. (911)

117. (number adj2 (nights or days)).ti,ab. (1891)

118. (primary care adj (visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (360)

119. (surgery adj (visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (11)

120. (GP adj (access or uptake or visit$ or contact$ or attendance$ or admission$ or episode$)).ti,ab. (102)

121. ((uptake or access) adj (service$ or care or intervention$)).ti,ab. (331)

122. ((clinic or surgery or hospital or “accident and emergency”) adj2 (work-flow or work flow)).ti,ab. (1)

123. (consultation$ adj2 (time or length)).ti,ab. (252)

124. (hospitalisation$ or hospitalization$ or rehospitalisation$ or rehospitalization$ or re-hospitalisation$ or re-hospitalization$).ti,ab. (21,193)

125. (hrql or hrqol or h qol or hql or hqol).ti,ab. (3414)

126. (qaly$ or quality adjusted life or quality of life or life quality).tw. (45,441)

127. 102 or 103 or 104 or 105 or 106 or 107 or 108 or 109 or 110 or 111 or 112 or 113 or 114 or 115 or 116 or 117 or 118 or 119 or 120 or 121 or 122 or 123 or 124 or 125 or 126 (246,148)

128. 20 or 70 (1,167,628)

129. 128 and 77 and 101 and 127 (4155)

130. Cool Kids.ti,ab. (20)

131. Sweet talk.ti,ab. (5)

132. Timid to Tiger.ti,ab. (3)

133. “problem solving for life”.ti,ab. (25)

134. Incredible Years.ti,ab. (160)

135. Triple P.ti,ab. (231)

136. friends program$.ti,ab. (38)

137. 130 or 131 or 132 or 133 or 134 or 135 or 136 (474)

138. 129 or 137 (4620)

Search strategy