Background
Methods
Search strategy
Component | Description, inclusion/exclusion criteria, process |
---|---|
Population | Studies were included if self-management support was delivered to populations with one or more of the exemplar long-term conditions (asthma, chronic kidney disease, chronic obstructive pulmonary disease, dementia, depression, diabetes (Type 1 and 2), epilepsy, hypertension, inflammatory arthritis, irritable bowel syndrome, stroke, low back pain, progressive neurological disease) selected for study in our overview of the literature [22]. |
Intervention | We included any implementation intervention which focused on, or incorporated, strategies to support self-management, and which were delivered as part of routine clinical service. Self-management support search terms included ‘confidence’, ‘self-efficacy’, ‘responsib*’, ‘autonom*’, ‘educat*’, ‘knowledge’, ‘(peer or patient) ADJ1 (support or group)’ and ‘(lifestyle or occupational) ADJ1 (intervention* or modification* or therapy)’ as well as relevant MeSH terms. |
Comparator | Typically ‘usual care’, although definition of ‘usual care’ varied between trials. The nature of the control service was noted and accommodated within our analysis. |
Outcomes | Use of healthcare services (including unscheduled use of healthcare services and hospital admission rates), health outcomes (including symptom control, biological markers of disease), and process outcomes (ownership of action plans, attendance at education sessions) and intermediary outcomes (self-efficacy). |
Settings | Any healthcare setting: hospital (in-patient or out-patient), community or remote (for example, web based) settings. |
Study design | Implementation studies [14, 15], including a range of methodologies: population level randomised controlled trials, quasi-experimental prospective studies, retrospective controlled studies, interrupted time series, controlled before and after studies, case–control, uncontrolled before and after studies, and observational studies. |
Databases | MEDLINE (1980 onwards), EMBASE (1974 onwards), CINAHL (1982 onwards), PsychINFO, AMED (1985 onwards), BNI, Database of Abstracts of Reviews of Effects and ISI Proceedings (Web of Science). |
Manual searching | Patient Education and Counseling, Health Education and Behaviour and Health Education Research. |
Forward citations | A forward citation search was performed on all included papers using ISI Proceedings (Web of Science). The bibliographies of all eligible studies were scrutinised to identify additional possible studies. |
Unpublished and in progress studies | UK Clinical Research Network Study Portfolio (www.clinicaltrials.gov) and the Meta Register of Controlled Trials (www.controlled-trials.com) |
Other exclusion criteria | We excluded papers not published in English. |
Defining and identifying implementation studies
Screening of titles and abstracts
Full-text screening
Assessment of methodological quality
Outcomes
Extraction of data
Data analysis
Interpretation and end-of-project workshop
Results
Description of the studies
Study quality and weight of evidence
Overview of results
Study | Design, size and quality | Intervention | Outcomes | ||||
---|---|---|---|---|---|---|---|
Patient | Professional | Organisation | Health service utilisation | Disease control and QoL | Process | ||
Primarily professional training | |||||||
Cleland 2007 [39] UK Primary care | Cluster RCT. FU: 6m 13 practices: 629 adults with poorly controlled asthma, Quality score = 24 | None | Intervention: one 3-hour interactive seminar vs. control | None | Not assessed | Routine data: SABA use and steroid courses: NS Sub-group: QoL (miniAQLQ): I: 6.49 (95%CI 6.40 to 6.59) vs C: 6.33 (95%CI 6.23 to 6.44) P = 0.03 (less than MCID of 0.5) Asthma control: NS | Not assessed |
Homer 2005 [30] US Primary care | Cluster RCT. FU 12m 43 practices: 13,878 children with asthma Quality score = 18 | None | Three one-day group training + two additional sessions + biweekly conference calls | Intended implementation of CCM | Admissions and ED visits: no between group differences reported | Asthma attacks and exercise limitation: no between group differences reported | Ownership of PAAP: I: 54% vs C: 41% (but large baseline difference) Use of preventer medication: I: 38% vs C: 39% Use of ICS I: 15% vs C: 17% |
Primarily patient education | |||||||
Delaronde 2005 [32] US Managed Care Organisation | Preference RCT. FU 12 (‘opt-in’ ‘opt-out’ ‘probably’ group were randomised) 399 adults, Quality score = 20 | Six-minute nurse-led telephonic case management vs usual care | None | None | Physician office visits, emergency department visits, hospitalisations: NS | Sub-group: No significant difference in the change in QoL (I: 0.26 vs C: 0.12) and within group changes < the MCID | Ratio of preventer to reliever medication. Increase in intervention group (0.18) was greater than in the control group (0.09) P = 0.04. Increase in the ‘opt-in’ group was greater at 0.29 (P = 0.01) |
Vollmer 2006 [35] US Managed Care Organisation | RCT, 6,948 adults, (192 had live calls) Quality score = 18 | Three 10-minute automated calls providing asthma review and personalised feedback | None | Provided as a service by the MCO | No between group difference in admissions/ED visits (% patients I: 4.1% vs C: 4.0% P = 0.88) or other unscheduled care | Asthma control: No difference in QoL (miniAQLQ I: 5.2 (SD 1.2) vs C: 5.1 (SD 1.2) P = 0.48) or any measure of asthma control | Medication use: No difference in ICS (% using ≥6 canisters/year I: 30.4% vs C: 29.8% P = 0.60) |
Bunting 2006 [31] US Managed Care Organisation | Repeated measures study, eight years of routine data 207 adults, Quality score=17 | One-to-one education + PAAP by a hospital based asthma educator. Sessions lasted 60 to 90 minutes + regular follow-up for five years by pharmacists. | None | Pharmacist and medication costs reimbursed by health plans. | From insurance claims: ED visits or hospitalisations /100 patients/y were lower during the programme (5.4, 2.6, 1.9, 5.4, 0) than in three years before (21.3, 22.2, 22.3) | Compared to baseline, at most recent follow up reduced: | PAAP ownership increased from 63% at baseline to 99% at follow-up (P <.0001) |
• % severe /moderate asthma B: 77% vs FU: 49% P <0.001 | |||||||
• working days lost B: 2.5/patient/year vs FU 0.5/patient/year | |||||||
Forshee 1998 [33] US Managed Care Organisation | Before and after study over 24 weeks 201 adults/children with poorly controlled asthma, Quality score = 15 | Tailored individualised education + videos + handouts | Nurse champions were educated about asthma | None | Compared to baseline, at follow up patients had: | Compared to baseline, at follow up patients had: | Monthly reviews, knowledge and confidence (non-validated questionnaire) increased significantly for both adults and children |
• Fewer episodes of unscheduled care (P ≤0.01) | • Improved severity classification (P <0.001) | ||||||
• Improved QoL (P ≤0.001) | |||||||
• Fewer days off work B: 6.5 vs FU: 3.9 (P <0.05) | |||||||
Gerald 2006 [34] Inner city elementary schools | Cluster RCT, 54 schools, 736 children, Quality score = 18 | 6 × 30 minute group education sessions for pupils with asthma + a clinical assessment with a paediatric allergist who developed a PAAP | None | Asthma education was provided for school staff A 30 minute classroom lesson was given to all children in grades I to IV in the school | Compared to control, intervention children had no difference in: | Compared to control, intervention children had: | Compared to control, school education resulted in a statistically significant increase in knowledge (P <0.0001) in 17 of the 18 schools |
• ED visits/child I: 0.09 (SD 0.28) vs C: 0.10 (SD 0.31) | • No difference in absenteeism : 3.88 days/child/year (SD 3.5) vs C: 3.21 (SD 3.2). | ||||||
• Admissions/child | |||||||
• d: 0.04 (SD 0.19) vs C: 0.02 (SD 0.14) | |||||||
Chini 2011 [47] Italy Primary schools | Before-and-after 2,765 children: 135 with asthma, Quality score = 15 | Clinical assessment and were given a PAAP with FU review at end of the year. Age-appropriate groups taught cognitive and breathing techniques | None | Lessons aimed at teachers, school personnel, parents, and schoolchildren to improve their knowledge of asthma | Not assessed | At the end of the year improved: | Not assessed |
• PedsQL: B: 2.2 (SD 0.79) vs FU: 3.5 (SD 0.73) P <0.001 | |||||||
• Parents’ perception of child’s QOL B: 3.1 (SD 0.6) vs FU: 3.5 (SD 0.4) P = 0.004 | |||||||
• Asthma symptoms (P <0.001) | |||||||
Primarily organisational change | |||||||
Kemple 2003 [40] UK Primary care | RCT, 545 adults, Quality score = 20 | None | None | Organisational intervention enclosing PAAPs (blank=I (AAP) or personalised= I (PAAP)) with invitations to review | There were no significant differences in admissions or out-of-hours consultations over the subsequent 12 months | There were no significant differences in prescriptions of short-acting beta2 agonists, peak flow, steroid courses | Compared to control OR of a review (95%CI): I (AAP): OR 1.92 (1.18 to 3.11); I (PAAP): OR 2.33 (1.37 to 3.93) |
Sub-group: Compared to control, OR of changing RCP3Qs score: I (AAP): OR 1.43 (0.80 to 2.56); I (PAAP): OR 1.46 (0.81 to 2.61) | Sub-group: Compared to control OR of understanding of self-management (95%CI): I (AAP): OR 1.28 (0.66 to 2.45); I (PAAP): OR 2.20 (1.13 to 4.30) | ||||||
Pinnock 2007 [41] UK Primary care | Controlled implementation trial, 1,809 adults and children, Quality score = 21 | Usual asthma review, including provision (or review) of self-management (with PAAP). | Existing practice asthma nurses who already had an accredited diploma on asthma care | Three reminders to patients due a review, with an option to book a telephone or face-to-face review. Opportunistic telephone calls to non-responders. | Not assessed | Sub-group: Compared to the control group, patients in the TC-option group had | More patients reviewed (I: 66.4% vs C: 53.8% risk difference 12.6% (95% CI 7.2 to 17.9)) |
• no difference in asthma control (ACQ mean (SD): I: 1.20 (1.00) vs C: 1.33 (1.13) mean diff 0.12 (−0.06 to 0.31) | Sub-group: Patients in the TC option group had greater: | ||||||
• enablement: P = 0.03 | |||||||
• no difference in asthma QoL | • confidence managing asthma (P = 0.007). | ||||||
Lindberg 2002 [48] Sweden Primary care | Cross-sectional survey, 8 practices: 347 adults + random sample of 20/practice for survey Quality score = 16 | The ANP provided regular review, including patient asthma education including a PAAP. | The Asthma Nurse Practitioner (ANP) had specialist asthma training. | With the exception of emergency visits and the yearly follow-up visit to their physician all visits were made to the asthma nurse | Patients from ANP centre had: | Survey (non-validated) Patients from ANP centre were less likely to |
Clinical records
|
• No difference in hospitalisations (I: 2.2% vs C: 3.7% NS) | • wake at night (P <0.01) | ANP centre was: | |||||
• Lower proportion of consultations (I: 43% vs C: 56% P <0.05) | • have activity limitation (P < 0.05) | • More likely record PF | |||||
• 18% lower total healthcare costs. | • have ≥2 asthma attacks in 6m (P <0.05) | • Discuss smoking | |||||
ANP centre patients had: |
Survey (non-validated)
| ||||||
• No difference in health status (EQ5D) | ANP centre patients were more likely to: | ||||||
• Increased sick leave. | • own PAAP (P <0.001) | ||||||
• use a PF meter | |||||||
• have knowledge about asthma (P <0.001) | |||||||
A whole systems approach | |||||||
Haahtela 2006 [45] Finland Primary, secondary and community settings | 10 year ITS, Population of Finland, Quality score = 10 (Note: many of the criteria did not apply) | Patient organisations arranged direct patient counselling and distributing information and resources free of charge | Education was provided for 5,300 respiratory specialists, 3,700 primary/secondary care professionals, 25,500 other healthcare professionals, 695 pharmacists | The Finnish Ministry of Social Affairs and Health recognised asthma as an important public health issue and set up the national programme | Over the 10 year programme: | Over the 10 year programme: | Over the 10 year programme: |
• Admissions fell from 110,000 to 51,000/year | • Sick leave decreased (from 2966 to 1920) | • Diagnosed asthmatics increased (from 225,000 to 350,000) | |||||
• Deaths fell from 123 to 85/year | • Number of people with asthma receiving disability payments decreased from 7212 to 1741 | • Proportion using ICS increased (33% to 85%) | |||||
• ED visits fell | • Deaths fell from 123/year to 85/year | • Smoking levels remained constant, | |||||
• Costs fell (from €1611 to €1031 per patient) | |||||||
Kauppi 2012 [46] | This publication reports follow on data from the Haahtela Finnish study (see previous entry). All the descriptive information is therefore the same. | In the six years after the end of the programme | In the three years after the end of the programme | ||||
• Admissions have continued to fall (from 32,000 hospital days 15,000 hospital days) | • Prevalence of asthma has continued to rise (from 6.8% to 9.4%) | ||||||
Souza-Machado 2010 [44] Brazil Community | Controlled implementation study over nine years, Population of Salvador and Recife (control city), Quality score = 11 (Note: many of the criteria did not apply) | Patient training: individual asthma education + monthly group sessions discussing asthma prevention and treatment | 512 primary healthcare physicians, nurses, pharmacists, social workers and managers were trained on asthma and rhinitis | Healthcare community project. Centres offered specialist care and free medication to patients with severe asthma | At nine years: | Over the nine years: in-hospital mortality decreased from 23 deaths in 2003 to one in 2006. (In Recife the in-hospital mortality rate increased from five deaths in 2003 to 6 in 2006) | From 2003 to 2006, the programme dispensed 220,889 units of inhaled medication for asthma control. There was a strong inverse correlation between hospitalisation rates and drug dispensation |
• Hospitalisation rates per 10,000 inhabitants at nine years: Salvador: 2.25 vs Recife 17.06 | |||||||
• The decline (2003 to 2006) was greater in Salvador (−74.2%) than Recife (−22.2%) P<0.001 | |||||||
Andrade 2010 [43] Brazil Primary healthcare network | Before and after study, 582 children (470 cases and 112 controls) Quality score = 19 | Individual and group educational activities, including PAAP | Patient education provided by pharmacists and health workers but no details of their training. | Healthcare community project. Free medication | At 12 months 5% of cases compared to 34% of controls had unscheduled asthma consultations P <0.01. | Not assessed | The use of ICS was greater in cases (67%) than controls (not given). All cases (users of the service) had a PAAP |
Bunik 2011 [38] US Secondary care paediatric clinics | Five year repeated measures study, 1,797 children clinic attendees, Quality score = 15 | Asthma educators provided education about medications and provided PAAPs. Telephone FU two weeks after unscheduled care | Monthly education sessions for junior medical staff and nurses. Computer and paper prompts to facilitate structured review with PAAPs | Pre-consultation questionnaires for families, templates for asthma reviews, respiratory therapist support for providing education and PAAPs. | There was no significant change in the proportion of children with ED visits (B:6% vs FU:6%) and hospitalisations (B:3% vs FU:3%) from 2006 to 2009. | Not assessed | Children seen three years after the intervention were more likely to: |
• Be given a PAAP (aRR 2.86 (95% CI 2.60–3.20) | |||||||
• Have an assessment of severity (aRR 1.47 (95% 1.41 to 1.54) | |||||||
• Be prescribed ICS (aRR 1.11 (95% CI 1.05 to 1.19) | |||||||
Swanson 2000 [42] Scotland Primary Care | Retrospective comparator study, 400 adults and children, Quality score = 16 | Asthma self-management education in asthma clinic | Professional training in implementing the BTS asthma guideline | Provision of paper-based templates | Compared to baseline, at follow-up patients in intervention practices were less likely to have had an ED attendance (p<0.05) or unscheduled consultation (p<0.05) | Compared to patients in control practices, attendees at intervention practice clinics reported greater improvements in asthma symptoms (p<0.001) | Compared to control practices, at FU patients in intervention practices were more likely to: |
• have and follow a PAAP (P <0.01) | |||||||
• have attended a review (P <0.05) | |||||||
Findley 2011 [37] US Community day care centres for pre-school children | Before-and-after study 35 centres, 1,908 children and their families, Quality score = 17 | Parents received asthma education from parent mentors and a PAAP, and were encouraged to talk with their child’s physician. Children played activities and games on asthma triggers | Professionals of children enrolled in the programme were offered. Physician Asthma Care Education (PACE) training | The centre staff received training on asthma and asthma management (including creating an ‘asthma-friendly centre’), identifying children with asthma, arranging a PAAP and handling emergencies | At 9 to 12 months the proportion of children with: | At 9 to 12 months the proportion of children with: | At 9-12 months: |
• Hospitalisations fell from 24% to 11% (P <0.001) | • Day-care absences reduced (56% to 38%) | • PAAP use increased from 47% to 70% | |||||
No ED visits increased from 25% to 53% (P <0.001). | • No night-symptoms increased (19% to 52%) (P <0.001) | • Staff knowledge increased 49% to 82% | |||||
• No day symptoms: increased ( 22% to 59%) (P <0.001) | • Parents’ knowledge increased 62 to 79%; | ||||||
• Parents’ confidence increased from 57% to 81% (P <0.001); | |||||||
Polivka 2011 [38] US Deprived community | Before-and-and after study, 243 children and their families, Quality score = 18 | Environmental assessment home repairs, educational home visits to reduce asthma triggers, and provide asthma education and PAAPs | Professionals completed the National Center for Healthy Homes practitioners’ course and an asthma educator course. | Costs included repair work, contractors, supplies for assessment and education provided to participants | At two years children had: | At two years children had fewer: | At two year follow up: |
• fewer emergency consultations (P <0.001)] | • day and night symptoms P <0.001 | • PAAP ownership increased B: 44% vs FU: 67% P = 0.007 | |||||
• no difference in admissions P = 0.229 | • days with activity limitation (P <0.001)] | • asthma knowledge increased (P <0.001) | |||||
• mean days off school B: 5.3 (SD 9.2) vs FU: 1.4 (SD 2.7) P <0.001 | • Caregiver | ||||||
• self-efficacy increased (P <0.001) |
Practical lessons from the authors’ reflections on the process of implementing complex self-management support interventions in routine clinical care. | |
• | |
• | Only a proportion of people accept the offer of self-management education, and all studies reported an attrition rate. For many interventions, especially those delivered in deprived communities, recruiting and retaining patients was a major challenge [37, 38]. Financial incentives (free access to care, free prescriptions, favourable insurance premiums, free patient resources) were potential strategies for increasing engagement [31, 37, 38, 43‐45]. |
• | |
• | Achieving change is a challenge, even in well-motivated teams [30]. There is a need to support professionals as they integrate new behaviour into practice [39]. Promising approaches include collaboratives, and plan/do/study/act (PDSA) cycles [30, 36], and introduction of self-management support as a component of improved chronic care [31, 43‐46]. |
• | |
• | Professional training in supporting self-management [36, 37, 45, 46], collaborative multidisciplinary working [36, 45], with good communication and referral systems between professionals [44], and involving existing staff members in the design and implementation of interventions [33, 36, 42] are potentially important ingredients of implementing self-management support. |
• | |
• | |
• | |
• | Technological solutions (such as computerised cognitive behaviour therapy programmes, automated telephone calls) are being explored and show some promise [35]. |