Systematic review on the cost-effectiveness of self-management education programme for type 2 diabetes mellitus

Accepted Manuscript Review Systematic review on the cost-effectiveness of self-management education programme for Type 2 diabetes mellitus J.X. Lian, S.M. McGhee, J. Chau, Carlos K.H. Wong, Cindy L.K. Lam, William C.W. Wong PII: DOI: Reference:

S0168-8227(16)30523-X http://dx.doi.org/10.1016/j.diabres.2017.02.021 DIAB 6877

To appear in:

Diabetes Research and Clinical Practice

Received Date: Accepted Date:

7 September 2016 14 February 2017

Please cite this article as: J.X. Lian, S.M. McGhee, J. Chau, C.K.H. Wong, C.L.K. Lam, W.C.W. Wong, Systematic review on the cost-effectiveness of self-management education programme for Type 2 diabetes mellitus, Diabetes Research and Clinical Practice (2017), doi: http://dx.doi.org/10.1016/j.diabres.2017.02.021

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Systematic review on the cost-effectiveness of self-management education programme

Systematic review on the cost-effectiveness of self-management education programme for Type 2 diabetes mellitus JX Lian1, SM McGhee2, J Chau2, Carlos KH Wong1, Cindy LK Lam1, William CW Wong1, 1

Department of Family Medicine and Primary Care, The University of Hong Kong,

Hong Kong 2

School of Public Health, The University of Hong Kong, Hong Kong

Running head: Systematic review on the cost-effectiveness of self-management education programme Corresponding author: William CW Wong Email address: [email protected] World count: 3770 Funding sources: This work was supported by Commissioned Study on Enhanced Primary Care (ref: EPC-HKU-1B & EPC-HKU-2). Conflicts of interest:

none declared.

1

Systematic review on the cost-effectiveness of self-management education programme

Abstract Objectives: A review of cost-effectiveness studies on self-management education programmes for Type 2 diabetes mellitus. Methods: Cochrane, PubMed and PsycINFO databases were searched for papers published from January 2003 through September 2015. Further hand searching using the reference lists of included papers was carried out. Results: In total, 777 papers were identified and 12 papers were finally included. We found eight programmes whose effectiveness analyses were based on randomised controlled trials and whose costs were comprehensively estimated from the stated perspective. Among these eight, four studies showed a cost per unit reduction in clinical risk factors (HbA1c or BMI) of US$491 to US$7,723 or cost per glycaemic symptom day avoided of US$39. In three studies the cost per QALY gained, as estimated from a life-time model, was less than US$50,000. However, one study found the programme was not cost-effective despite a gain in QALYs at the one-year follow up. Conclusion: A small number of cost-effectiveness studies were identified with only eight of sufficiently good quality. The cost of a self-management education programme achieving reduction in clinical risk factors seems to be modest and is likely to be cost-effective in the long-term.

2

Systematic review on the cost-effectiveness of self-management education programme

INTRODUCTION According to the World Health Organization, it is estimated that around 9% of adults suffer from diabetes mellitus (DM), equivalent to 347 million people around the world.[1] With the numbers predicted to increase and the known risk of micro- and macrovascular complications, this imposes an increasing disease and economic burden on individuals and society. For a long time, DM self-management has been important to prevent the onset and delay progression of complications so those with DM already often take much of the responsibility for their day-to-day care. The encouragement of this self-empowerment can take a variety of forms and include different components but essentially involves education to facilitate acquisition of knowledge, skill, and abilities for diabetes self-care.[2]

Systematic reviews of

randomised controlled trials (RCT) have shown that self-management education programmes can lead to reductions in HbA1c, [3-6] blood pressure[5] and lipids[3] and to improvement in self-management skills and self-efficacy.[5] The American Diabetes Association (ADA) stated in 2015 that all individuals with diabetes should receive diabetes self-management education or support at diagnosis and as needed thereafter.[2]

Even with this evidence on effectiveness, whether self-management education programmes represent an efficient use of resources remains an open question. A review by Loveman et al. on the cost-effectiveness of such programmes only found two studies, both from the US, between 1980 and 2003.[7] One of these found that a 3

Systematic review on the cost-effectiveness of self-management education programme

programme focused on improving diet and exercise was associated with 0.092 units gain in well-being at follow up and cost US$10,870 per well-year compared with an education-only

control

group.[8]

The

other

study

found

that

a

dietary

self-management programme cost US$62 to $105 per 1% reduction in dietary fat or saturated fat and US$8 per 1 mg/dl reduction in serum cholesterol when compared with usual care alone.[9] There has been a recent upsurge of interest in the management of chronic disease and in the development of self-management education programmes. We aim here to summarize the current evidence on cost-effectiveness of self-management education programmes for people with DM.

METHODS Search strategy We searched the databases relevant to public health and health promotion interventions including the Cochrane Database, PubMed and PsycINFO using alternative terms for key words which were: Indicating cost-effectiveness: “economic evaluation” OR “cost effectiveness” OR “cost consequence analysis” OR “cost utility” OR “cost benefit” OR (“technology assessment” AND “biomedical”)

OR “health technology assessment” OR

“quality-adjusted life-year” OR “QALY”; Indicating type 2 diabetes: “type II diabetes” OR “non insulin dependent diabetes” OR (“diabetes mellitus” AND “type 2”) OR “NIDDM” OR “T2DM” OR “type 2 diabetes”;

4

Systematic review on the cost-effectiveness of self-management education programme

Indicating self-management education programme: “empowerment” OR “education” OR “health education” OR “knowledge improvement” OR (behavio* AND “change”) OR “lifestyle modification” OR “self management” OR “self care” OR (behavio* AND “therapy”) OR “psychotherapy” OR “self help group” OR “DSME” OR “DSMT” OR “DSMI” OR “DSMA”. The detailed search terms used are listed in Appendix 1. In addition, a hand search was conducted using the reference lists of the included studies. The literature search was first conducted in 2013 with a search period from 1st January 2003 to 13th September 2013 and was updated in 2015 to 18th September 2015.

Studies which fulfilled all the following criteria were included: 1) RCT or controlled clinical trial that compared an education/self-management programme, targeting multiple

cardiovascular

risk

factors,

with

usual

care

or

with

another

education/self-management programme, conducted in patients with type 2 diabetes; 2) Published between 1st January 2003 and 18th September 2015 (inclusive); 3) Published in English; and, 4) Reported on the cost-effectiveness of the intervention programme. Criteria for exclusion were: a. non-English publication; b. study not related to diabetes (e.g. liver disease); c. study of type 1 diabetes mellitus or gestational diabetes mellitus only; d. study of diabetes screening or prevention; 5

Systematic review on the cost-effectiveness of self-management education programme

e. study not related to patient education, e.g., pharmacological treatment, nutrition supplement or gastric bypass surgery; f. study not specific for diabetes e.g., education for patients with chronic disease; g. study of education interventions specific for managing one single diabetic complication or comorbidity e.g., foot ulcer; h. systematic review or meta-analyses; i.

study with no comparison or control group;

j.

study which did not report on cost-effectiveness of intervention programme;

Two independent reviewers screened titles and abstracts of all identified publications, as well as the full text assessment. Any discrepancy between the two reviewers was sent to a third independent reviewer for the final decision.

Quality assessment and Data extraction A published checklist was used to assess the quality of the included studies. For quality assessment we used Drummond’s guideline which is a comprehensive assessment of cost-effectiveness analysis including aspects of study design, data collection, analysis and interpretation of results.[10] This guideline is the basis of many other published checklists and is recommended by the Cochrane Library.[11] We extracted data from each study into tables including authors, publication year, country, study population, programme description, effectiveness data, perspective, types of costs, type of model if used, main assumptions and cost-effectiveness results. Quality assessment was done by two independent reviewers. Any discrepancy 6

Systematic review on the cost-effectiveness of self-management education programme

between the reviewers was settled by discussion. Data extraction was undertaken by one reviewer. We summarized the studies under the following 3 sections: effectiveness of the programme, costs and cost-effectiveness of the programme.

RESULTS Study selection Our initial search identified 942 records, reduced to 777 after removing duplicates. The title and abstract screen yielded 46 for full-text assessment. A final 11 studies met our inclusion criteria and were reviewed.[12-22] Searches of reference lists of these included studies identified one further relevant study[23] so 12 studies were included in the full review. The results of the literature search are presented in a PRISMA flowchart (Figure 1).

Quality assessment The studies ranged between meeting 58% of quality criteria [13] to 97% of criteria [15, 18] and, although no study achieved 100%, two were over 95% (Table 1). The commonest unmet criteria were in the Data Collection section including quantities of resources not being reported separately from their unit costs (7/12), methods to value health states and other benefits not stated (4/8), and details of the subjects from whom valuations were obtained not given (4/8). Some criteria in the Analysis and Interpretation of Results section were not met by several papers including the choice of discount rate(s) not justified (3/6), and details of statistical tests and confidence 7

Systematic review on the cost-effectiveness of self-management education programme

intervals not given for stochastic data (8/12).

Description of the studies Only two studies were from developing countries[18, 22] and no studies were on an Asian population (Table 2). Five studies targeted vulnerable populations with poor glycaemic control, underserved minority groups or postmenopausal women with type 2 diabetes.[14, 16, 17, 21, 23] The programmes were implemented in the primary care setting in eight studies,[13-16, 19, 21-23] in secondary or tertiary care settings in two[12, 18] and two did not specify. The educational input in terms of components and educators varied across the studies with group education used in five[15, 16, 20, 22, 23], one-to-one education in two[19, 21] and telephone calls in three[13, 14, 17] while two studies did not mention the form of education.[12, 18] The programmes were led by trained educators i.e. health care educator, health promoter, health worker or trained telecarer in five studies[13, 15, 17, 21, 22], pharmacists in two,[18, 19] allied health professionals i.e. dietician, exercise physiologist and stress management specialist in two,[16, 23] and not specified in the remaining three.[12, 14, 20]

Effectiveness of the programme Most of the effectiveness findings used in the CEA studies (11/12) were derived from RCTs and so would be considered as good evidence (Table 2). The length of follow up was either 6- or 12-months. All of these studies found that the programme examined had an effect in one or more of the following outcome measurements, 8

Systematic review on the cost-effectiveness of self-management education programme

reduction in HbA1c,[12, 13, 16, 17, 20, 21, 23] reduction in BMI, [15, 16, 23] reduction in blood pressure,[22] reduction in smoking,[15] gain in QALYs,[14, 18] or reduction in days with symptoms of hyperglycaemia or hypoglycaemia[19] when compared with the control group. The remaining study collected data from participants attending five programmes with a self-management component and was compared with data from a comparison group of non-participants but randomisation was not mentioned so the sample may not have been randomly selected and this may have introduced bias relative to the controls. [20] The five programmes had different components and targeted different subjects and so the effectiveness was a mixed effect from the groups of subjects with different characteristics. The evidence of effectiveness from this study was considered relatively weak.

Cost The key elements in estimating the cost in each study are summarized in Table 3. The perspective taken for the costing is important in determining which cost items should be included. The most common perspective adopted in these studies was the provider perspective which usually resulted in the inclusion of only direct costs of setting up and running the programme (6/12).[14, 15, 17, 19, 20, 23] Five of the 6 studies taking this perspective documented comprehensive costs[14, 15, 17, 19, 23] which included different types of staff time, educational materials and, in some cases venue rental, equipment and telephone charges. Only three studies adopted a societal perspective 9

Systematic review on the cost-effectiveness of self-management education programme

and included, not only direct programme costs but also subjects’ own costs i.e. time cost and travel cost incurred in participating in the programme.[16, 21, 22] The costs in these three studies were comprehensive although one study did not include subject travel cost owing to the programme serving only a small geographic area.[21]

One

study took the patient’s perspective, who in this case was the health care purchaser, and included only the direct costs of the self-management equipment, materials and health care as purchased by the patient.[18] However this costing could be considered not comprehensive because the patient’s time and travel cost were not included. We could not judge whether the costs were comprehensively estimated in the remaining three studies because they either did not describe the components of the programme’s cost in sufficient detail [20] or did not state which perspective had been taken. [12, 13]

Cost-effectiveness of self-management education programmes The cost-effectiveness results are summarised in Table 3. The quality and usefulness of these results largely depend on whether the costs were estimated comprehensively and effectiveness was derived from good evidence. We found eight studies with good evidence on both cost and effectiveness among which five applied the effectiveness data from the RCT to calculate an incremental cost-effectiveness ratio (ICER) without modelling. Ritzwoller et al. found that a Mediterranean lifestyle programme for postmenopausal women with diabetes had an ICER, from the provider’s perspective, of US$3,808 per incremental change (%) in 10

Systematic review on the cost-effectiveness of self-management education programme

HbA1c, US$2,354 per unit (kg/m2) reduction in BMI , US$644 per unit improvement in the Problem Areas in Diabetes self-care summary score which was a quality-of-life assessment, and US$196 per gramme reduction in intake of saturated fatty acids when compared to usual care.[23] The same authors conducted a later study of a similar programme but culturally adapted to Latina women with type 2 diabetes and found an ICER of US$7,723 per unit (%) reduction in HbA1c and US$7,723 per unit (kg/m2) reduction in BMI compared with usual care.[16] They commented that, although their costs may be higher than interventions that address only one risk factor, the potential costs of long-term health care are high for this group and so long term cost-effectiveness may be even better than the short-term results presented here. The study by Schechter et al. of a telephone-based self-management programme for lower-income adults with type 2 diabetes and a baseline HbA1c of ≥7.5% gave ICERs, from the provider’s perspective, of US$490.6 per unit (%) reduction in HbA1c and of US$2,617.4 per extra person achieving a target HbA1c of <7% compared with a control group receiving only diabetes self-management printed materials.[17] They concluded that a modest improvement in HbA1c can be achieved at a moderate cost using a telephone-based intervention. Hendrie et al. found that a pharmacist-led diabetes management education programme with tailored individualized education was effective in reducing glycaemic-symptom days

with

an

ICER,

from

the

provider’s

perspective,

of

US$39

per

glycaemic-symptom day avoid compared with standard pharmacy care.[19] This was lower than the subjects’ stated willingness to pay to avoid a hypoglycaemic day, 11

Systematic review on the cost-effectiveness of self-management education programme

which was valued at US$73. Another telephone-based study by Handley et al. of automated support with nurse backup for low income patients with poor HbA1c control (≥8%) identified a gain in QALYs of 0.012 over one year compared to usual care. This gave an ICER, from the provider’s perspective, of U$65,167 per extra QALY gained when both the start-up and continuing costs were included but US$32,333 per QALY when only the continuing cost was included.[14] At the threshold of US$50,000 per QALY gained which is widely adopted in US, this programme was not considered cost-effective when set up costs were included. The other three CEA studies with good evidence of cost and effectiveness used a cost-effectiveness model applying the effectiveness findings from the RCT into the model to simulate the development of diabetic complications and utilization of health service with a long-term time horizon (Table 4). Assumptions are commonly made in such a model and would have direct impact on the simulation results.

Gillett et al.

included only newly diagnosed cases in the DESMOND study [15] and modelled costs and benefits over a life time. The 12-month effectiveness results from the RCT were incorporated into the Sheffield type 2 diabetes model [24] which is an individual level model and simulates the progression of five major diabetic complications i.e. retinopathy, nephropathy, neuropathy, coronary heart disease and cerebrovascular disease using prediction algorithms and data on key risk factors including HbA1c, blood pressure and smoking status. DESMOND resulted in a mean lifetime gain of 0.0392 QALYs per person and an ICER, from the provider perspective, of £5,387 per QALY gained based on the costs measured in the trial and of £2,092 per QALY gained 12

Systematic review on the cost-effectiveness of self-management education programme

based on an estimate of real world costs i.e. a programme for a hypothetical primary care trust with an average population of 329,550 patients. This programme would be cost-effective under a cost per QALY threshold of £30,000/QALY. The lifetime gain in QALYs was mainly due to a reduction in smoking rates and weight and also improvements in plasma lipid levels leading to reduction in complications and improved survival. Their result was robust to changing assumptions of the time over which the programme effects would last i.e. 3 years after the end of the RCT (main model) down to only 1 year and a lesser effect of smoking on mortality from other causes i.e. hazard ratio of 1.24 vs 1.36 (main model). Prezio et al. compared a community-based diabetes education programme tailored to individuals and delivered by community health workers to uninsured Mexican Americans and found a significant reduction in HbA1c compared to usual care. These results were incorporated into an individual model called Archimedes[25] which modelled five major diabetic complications and resulted in an average of 0.056 QALYs gained per person over 20 years with an ICER, from a societal perspective, of US$355 per QALY gained compared with usual care.[21] It was not stated how long the programme’s effects were expected to last in the model, but since the programme was designed to be continued after the first year, the effect was likely assumed to be persistent. When the effectiveness of the programme (i.e. reduction in HbA1c) was decreased by 30% from the base-case scenario (main model), the programme failed to be cost-effective with an ICER exceeding US$50,000. How long the short-term effectiveness, i.e. actual reduction in HbA1c or change in lifestyle habits found in the 13

Systematic review on the cost-effectiveness of self-management education programme

RCT, can be sustained over time is an important parameter which can substantially affect the long term benefits and cost-effectiveness. Mash et al. evaluated a group education programme delivered by health promoters which had shown a reduction in blood pressure in the RCT compared to usual care when offered to underserved communities in South Africa.[22] The effectiveness results were incorporated into a Markov model but only simulated cardiovascular complications without considering microvasular complications. The modelled results showed a gain of 0.067 QALYs per subject over their lifetime with an ICER, from the societal perspective, of US$1,862 per QALY gained. The authors assumed that the effect of the programme would persist throughout life and that they would incur an annual intervention cost due to follow-up education. They tested a variety of assumptions including one year of costs but with persistent benefits, one year of costs with one year of benefits, or one year of costs with three years of declining benefits which all improved the programme’s cost-effectiveness to the extent that it became cost-saving. Three other CEA studies had good effectiveness data from the RCT but were relatively weak in the costing. The first was a monthly educational intervention by pharmacists for subjects in a tertiary hospital which achieved a gain of 0.12 QALYs per subject over a year compared with usual care and resulted in an ICER, from the patient’s perspective, of US$571 per QALY gained.[18] The estimation of cost did not include the subject’s time and travel costs so may have underestimated the costs but it did find relatively low costs per QALY so adding these costs may still have resulted in 14

Systematic review on the cost-effectiveness of self-management education programme

a cost-effective programme. The second study modelled a patient-centred strategy with a lifetime ICER, compared with usual care, of €16,353 per QALY gained.[12] The perspective taken for the costs was not stated but could be guessed as being a provider perspective. The third study was on patient education with telemedicine support aimed at improving glycaemic control and the effectiveness data derived from the RCT was used to simulate lifetime costs and benefits with a resulting cost-effectiveness ratio of £43,400 per QALY gained under trial conditions and £33,700 per QALY gained in routine care with a full caseload of subjects with HbA1c≥7%.[13] This study also did not state the perspective taken for the costs but could also be guessed to be a provider perspective. The final CEA study reported on the Alliance programmes [20] and, from the provider perspective, they were not considered cost-effective with an ICER of US$61,011 per QALY. This is the only study in which the effectiveness data was not derived from an RCT and the costs were not reported in detail so it is difficult to draw further conclusions from this study.

DISCUSSION This review identified 12 cost-effectiveness studies of which eight had good data on both effectiveness and costs and were used for the main findings of this review. The cost of achieving a one-unit reduction in clinical risk factors e.g. HbA1c (%) and BMI (kg/m2) through a self-management education programme appears modest with a 15

Systematic review on the cost-effectiveness of self-management education programme

range from US$490 to US$7,723.[16, 17, 23] The improvement in clinical risk factors would reduce the risk of developing diabetic complications many years later. There is clearly the opportunity to benefit the health of these individuals and reduce costs in the long-term.

Modelling can be very helpful to

simulate how changes in risk

factors, that can be observed in the short-term studies, might affect long-term outcomes such as complications and deaths that would otherwise occur years later. However the model is only as good as its structure and parameters and this approach relies heavily on having good data on effectiveness and costs. The RCT findings described here are important because they can be used to predict, with a variety of assumptions, the longer-term complications and associated health care utilisation of the subjects. Among the eight, good quality, CEA studies three used modelling to project their results over the longer term and found that self-management education programmes were cost-effective over 20 years or a life time no matter whether examined from the provider or the societal perspective. Only one study which we classed as having good quality data did not find self-management programmes cost-effective.[14] This was Handley’s study which calculated an ICER based on the effectiveness (QALYs gained) and costs observed in the RCT over the short-term only, without modelling, and found it to be US$65,167, beyond the threshold of US$50,000. This may have been mainly due to a high set up cost so could improve with a longer term perspective. Among the 12 studies in this review, there were only three which adopted a societal perspective and estimated the subject’s cost as well as the programme costs. 16

Systematic review on the cost-effectiveness of self-management education programme

Self-management education programmes normally need quite a large commitment to attend educational sessions, which might be an opportunity cost for subjects, e.g. in taking time from work or leisure, and they may need to change lifestyle and habits. Those studies which failed to include subject costs might bias the findings towards cost-effectiveness by narrowing the scope of measured costs compared with actual costs. All of the included studies measured effectiveness of the self-management programme by changes in clinical risk factors or QALYs during the follow up period. Observation of diabetic complications would have been difficult due to the limited follow up time in all the RCTs. There are limitations in this review. Firstly, due to only a small of number of studies being finally included in this review, we were not able to analyse which type of self-management education programme is most cost-effective e.g. group-based or individual education, or by which type of educator. However, the answer to this question may depend on setting and context. Secondly, results from the reviewed studies cannot easily be compared because of the heterogeneous nature of the subjects’ characteristics, educational components, differing control groups and outcomes used. However, we have tried to summarize the main findings in a way which highlights the quality of the cost and effectiveness data and extracts the important information for those interested in applying this type of patient support. This detail is not always present in other reviews.

Thirdly, it is possible that publication bias in the past has

resulted in mainly studies with positive results being published and any studies with 17

Systematic review on the cost-effectiveness of self-management education programme

inconclusive or negative findings would therefore not have been included. However in recent years, it is likely that a good quality RCT would be published, even with negative findings. In conclusion, there is a limited number of CEA studies, as identified in this review, and eight of them were considered as being of sufficiently good quality to inform practice. To have robust results on cost-effectiveness, we must first have good data on effectiveness and comprehensive data on costs. Ideally a lifetime model using such data can predict the longer-term costs and benefits. The evidence from the reviewed studies showed that the cost of achieving reduction in clinical risk factors appears to be modest and self-management education programmes are likely to be cost-effective in the long-term.

FUNDING This work was supported by Commissioned Study on Enhanced Primary Care (ref: EPC-HKU-1B & EPC-HKU-2).

CONFLICTS OF INTEREST All authors have no conflicts of interest.

ACKNOWLEDGEMENT We would like thank Joe Lee, Winnie Wong for developing the search terms of the literature, screening and reviewing the title and abstract of the records. 18

Systematic review on the cost-effectiveness of self-management education programme

REFERENCE 1.

World

Health

Organization.

Diabetes

Programme.

Available

from:

http://www.who.int/diabetes/en/. Last accessed 6 Nov 2015 2.

Powers MA, Bardsley J, Cypress M, Duker P, Funnell MM, Hess Fischl A et al. Diabetes

Self-management Education and Support in Type 2 Diabetes: A Joint Position Statement of the American Diabetes Association, the American Association of Diabetes Educators, and the Academy of Nutrition and Dietetics. Diabetes Care. 2015 Jul;38(7):1372-82. 3.

Sherifali D, Bai JW, Kenny M, Warren R, Ali MU. Diabetes self-management programmes in

older adults: a systematic review and meta-analysis. Diabet Med. 2015 Nov;32(11):1404-14. 4.

Norris SL, Lau J, Smith SJ, Schmid CH, Engelgau MM. Self-management education for adults

with type 2 diabetes: a meta-analysis of the effect on glycemic control. Diabetes Care. 2002 Jul;25(7):1159-71. 5.

Steinsbekk A, Rygg LO, Lisulo M, Rise MB, Fretheim A. Group based diabetes self-management

education compared to routine treatment for people with type 2 diabetes mellitus. A systematic review with meta-analysis. BMC Health Serv Res. 2012;12:213. 6.

Ellis SE, Speroff T, Dittus RS, Brown A, Pichert JW, Elasy TA. Diabetes patient education: a

meta-analysis and meta-regression. Patient Educ Couns. 2004 Jan;52(1):97-105. 7.

Loveman E, Cave C, Green C, Royle P, Dunn N, Waugh N. The clinical and cost-effectiveness of

patient education models for diabetes: a systematic review and economic evaluation. Health Technol Assess. 2003;7(22):iii, 1-190. 8.

Kaplan RM AC, Wilson DK. The cost–utility of diet and exercise interventions in

non-insulindependent diabetes mellitus. Health Promot. 1987;2:331–40. 9.

Glasgow RE, La Chance PA, Toobert DJ, Brown J, Hampson SE, Riddle MC. Long-term effects

and costs of brief behavioural dietary intervention for patients with diabetes delivered from the medical office. Patient Educ Couns. 1997 Nov;32(3):175-84. 10. Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economic submissions to the BMJ. The BMJ Economic Evaluation Working Party. BMJ. 1996 Aug 3;313(7052):275-83. 11. Mathes T, Walgenbach M, Antoine SL, Pieper D, Eikermann M. Methods for systematic reviews of health economic evaluations: a systematic review, comparison, and synthesis of method literature. Med Decis Making. 2014 Oct;34(7):826-40. 12. Dijkstra RF, Niessen LW, Braspenning JC, Adang E, Grol RT. Patient-centred and professional-directed implementation strategies for diabetes guidelines: a cluster-randomized trial-based cost-effectiveness analysis. Diabet Med. 2006 Feb;23(2):164-70. 13. Mason JM, Young RJ, New JP, Gibson JM, Long AF, Gambling T et al. Economic Analysis of a Telemedicine Intervention to Improve Glycemic Control in Patients with Diabetes Mellitus: Illustration of a Novel Analytic Method. Disease Management & Health Outcomes. 2006;14(6):377-85. 14. Handley MA, Shumway M, Schillinger D. Cost-effectiveness of automated telephone self-management support with nurse care management among patients with diabetes. Annals of Family Medicine. 2008;6(6):512-8. 15. Gillett M, Dallosso HM, Dixon S, Brennan A, Carey ME, Campbell MJ et al. Delivering the 19

Systematic review on the cost-effectiveness of self-management education programme

diabetes education and self management for ongoing and newly diagnosed (DESMOND) programme for people with newly diagnosed type 2 diabetes: Cost effectiveness analysis. BMJ: British Medical Journal. 2010;341(7770):1-10. 16. Ritzwoller DP, Sukhanova AS, Glasgow RE, Strycker LA, King DK, Gaglio B et al. Intervention costs and cost-effectiveness for a multiple-risk-factor diabetes self-management trial for Latinas: economic analysis of inverted exclamation markViva Bien! Transl Behav Med. 2011 Sep 1;1(3):427-35. 17. Schechter CB, Cohen HW, Shmukler C, Walker EA. Intervention costs and cost-effectiveness of a successful

telephonic

intervention

to

promote

diabetes

control.

Diabetes

Care.

2012

Nov;35(11):2156-60. 18. Adibe MO, Aguwa CN, Ukwe CV. Cost-utility analysis of pharmaceutical care intervention versus usual care in management of nigerian patients with type 2 diabetes. Value in Health Regional Issues

[serial

on

the

Internet].

2013;

2(2):

Available

from:

http://onlinelibrary.wiley.com/o/cochrane/clcentral/articles/275/CN-00916275/frame.html. 19. Hendrie D, Miller TR, Woodman RJ, Hoti K, Hughes J. Cost-effectiveness of reducing glycaemic episodes through community pharmacy management of patients with type 2 diabetes mellitus. The Journal of Primary Prevention. 2014 2015-02-24;35(6):439-49. 20. Lewis MA, Bann CM, Karns SA, Hobbs CL, Holt S, Brenner J et al. Cross-site evaluation of the Alliance to Reduce Disparities in Diabetes: clinical and patient-reported outcomes. Health Promot Pract. 2014 Nov;15(2 Suppl):92s-102s. 21. Prezio EA, Pagán JA, Shuval K, Culica D. The Community Diabetes Education (CoDE) program: Cost-effectiveness and health outcomes. American Journal of Preventive Medicine. 2014 2015-06-24;47(6):771-9. 22. Mash R, Kroukamp R, Gaziano T, Levitt N. Cost-effectiveness of a diabetes group education program delivered by health promoters with a guiding style in underserved communities in Cape Town, South Africa. Patient Education and Counseling. 2015 2015-04-29;98(5):622-6. 23. Ritzwoller DP, Toobert D, Sukhanova A, Glasgow RE. Economic analysis of the Mediterranean Lifestyle

Program

for

postmenopausal

women

with

diabetes.

Diabetes

Educ.

2006

Sep-Oct;32(5):761-9. 24. Brennan A, Chick SE, Davies R. A taxonomy of model structures for economic evaluation of health technologies. Health Econ. 2006 Dec;15(12):1295-310. 25. Eddy DM, Schlessinger L. Archimedes: a trial-validated model of diabetes. Diabetes Care. 2003 Nov;26(11):3093-101.

TABLES Table 1. Quality assessment using Drummond’s checklist Table 2. Description of the studies Table 3. Summary of costs and cost-effectiveness Table 4 Summary of the modelling 20

Systematic review on the cost-effectiveness of self-management education programme

FIGURE LEGENDS Figure 1. PRISMA flowchart

21

24

Table 1. Quality assessment using Drummond’s checklist Item

Ritzwoller

Ritzwoller

Schechter

Handley

Gillett

Dijkstra

Mason

Adibe

Hendrie

Prezio

Mash

Lewis

[23]

[16]

[17]

[14]

[15]

[12]

[13]

[18]

[19]

[21]

[22]

[20]

(1) The research question is stated

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

(2) The economic importance of the

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

N

N

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

N

Y

Y

Study design

research question is stated (3) The viewpoint(s) of the analysis are clearly stated and justified (4) The rationale for choosing the alternative programs or interventions compared is stated (5) The alternatives being compared are clearly described (6) The form of economic evaluation used is stated (7) The choice of form of economic evaluation is justified in relation to the questions addressed Data collection (8) The source(s) of effectiveness estimates used are stated (9) Details of the design and results of effectiveness study are given (if based on a single study)

26

(10) Details of the method of

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

NA

NA

NA

Y

Y

Y

N

Y

NA

N

N

N

NA

NA

NA

Y

Y

Y

N

Y

NA

N

N

N

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

Y

NA

NA

NA

NA

NA

NA

NA

NA

N

Y

Y

N

Y

N

N

N

Y

N

Y

N

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

N

N

Y

Y

Y

N

Y

N

N

Y

Y

Y

Y

N

Y

N

N

Y

Y

Y

Y

Y

Y

Y

Y

N

synthesis or meta-analysis of estimates are given (If based on an overview of a number of effectiveness studies) (11) The primary outcome measure(s) for the economic evaluation are clearly stated (12) Methods to value health states and other benefits are stated (13) Details of the subjects from who valuations were obtained are given (14) Productivity changes (if included) are reported separately (15) The relevance of productivity changes to the study question is discussed (16) Quantities of resources are reported separately from their unit costs (17) Methods for the estimation of quantities and unit costs are described (18) Currency and price data are recorded (19) Details of currency of price

27

adjustments for inflation or currency conversion are given (20) Details of any model used are

NA

NA

NA

NA

Y

Y

Y

NA

NA

Y

Y

Y

NA

NA

NA

NA

Y

Y

Y

NA

NA

Y

Y

Y

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

(23) The discount rate(s) is stated

NA

NA

NA

NA

Y

Y

Y

Y

NA

Y

N

Y

(24) The choice of rate(s) is justified

NA

NA

NA

NA

Y

N

N

Y

NA

Y

NA

N

(25) An explanation is given if costs

NA

NA

Y

NA

NA

NA

NA

NA

NA

NA

N

NA

N

N

Y

N

Y

N

N

Y

N

N

N

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

N

Y

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

given (21) The choice of model used and the key parameters on which it is based are justified Analysis and interpretation of results (22) Time horizon of costs and benefits is stated

or benefits are not discounted (26) Details of statistical tests and confidence intervals are given for stochastic data (27) The approach to sensitivity analysis is given (28) The choice of variables for sensitivity analysis is justified (29) The ranges over which the variables are varied are stated (30) Relevant alternatives are compared

28

(31) Incremental analysis is reported

Y

Y

Y

Y

Y

Y

N

Y

Y

Y

Y

Y

(32) Major outcomes are presented

Y

Y

N

Y

Y

Y

N

Y

N

Y

N

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

no. of items met / no. of applicable

23/25

23/25

24/26

24/28

30/31

25/31

18/31

28/29

23/25

25/31

24/31

21/31

items (%)

(92%)

(92%)

(92%)

(86%)

(97%)

(81%)

(58%)

(97%)

(92%)

(81%)

(77%)

(68%)

in a disaggregated as well as aggregated form (33) The answer to the study question is given (34) Conclusions follow from the data reported (35) Conclusions are accompanied by the appropriate caveats

Y=yes; N=no; NA=not appropriate;

29

Table 2. Description of the studies Author;

Country

Year;

Study

Subjects;

Setting;

design

Length

Sampl

Program;

Control

Effectiveness data from

of FU

e size

Component;

group

the RCT

Mediterranean Lifestyle

Usual

At 6-month, the

Program:

care

intervention vs control

(month) Ritzwoller;

US

RCT

2006;[23]

Postmenopaus

Primary

al women with

care

6

Educator; 279

 Addressed diet, physical activity, stress management,

type 2 diabetes at risk for coronary heart

group  more reduction in HbA1c (0.34%)

social support and smoking;  By exercise specialist,

disease;

 more reduction in BMI (0.55 kg/m2)

dieticians, stress

 more reduction in saturated fatty acids

management specialist.

intake (6.62g)  more improvement in self-care summary score (2.01). Ritzwoller; 2011;[16]

US

RCT

Women Latinas

Primary

with type 2

care

6

280

diabetes;

Viva Bien Program:

Usual

At 6-month, the

 Culturally adapted

care

intervention vs control

“Mediterranean Diet”- style

group

eating plan, physical activity

 more reduction in HbA1c (0.6%)

and stress management regimens, group social support, and smoking

30

 more reduction in BMI(0.6kg/m2).

cessation;  By dietician and exercise physiologist. Schechter

US

RCT

2012;[17]

Urban adults

Not clear

12

526

Behavioural telephonic

Receive

At 12-month, the

with type 2

intervention

d only

intervention vs control:

diabetes with a



up to 10 self-management

the

baseline HbA1c

support telephone calls, one

diabetes

 more reduction in HbA1c (0.36%)

≥7.5% and with

call every 4-6 weeks to

self-man

 11.8% of the telephone

lower-income

discuss self-management, in

agemen

group achieved HbA1c

addition to the same print

t print

≤7% while 5.1% of the

diabetes self-management

material

control group.

education materials that

s

were sent to the control group 

Hendrie;

Australi

2014; [19]

a

RCT

Subjects

with

type 2 diabetes;

Primary

6

160

care

By a health educator

Pharmacist-led

Standar

At 6-month, the

d

intervention vs control:

Program (DMEP):

pharma

 a tailored individualized education program

cy care

 reduced 1.86 days with symptoms of

Management

Diabetes Education

hyperglycaemia and

containing up to 3 h of

hypoglycaemia per

one-on-one diabetes

patient per month.

education.  By pharmacist. Handley 2008;[14]

US

RCT

Subjects with

Primary

HbA1c

care

12

226

31

Automated telephone

Usual

At 12-month, the

self-management support with

care

intervention vs control:

value≥8%;

 incremental QALY gain of 0.012 per subject

nurse care management (ATSM):  uses interactive phone technology to provide surveillance, patient education and, when combined with nurse care management in the form of one-on-one counselling  not clear on the type of educator.

Gillett;

UK

RCT

2010;[15]

Newly

Primary

diagnosed type

care

12

824

2 diabetes

Diabetes education and self

Usual

At 12-month, the

management for ongoing and

care

intervention vs control:  more weight reduction of 1.26 kg per subject

newly diagnosed program:  A 6-hour structured group education program, focuses

 greater reduction in

on lifestyle factors, such as

smoking status

food choices and physical activity and cardiovascular risk factors;  By two professional healthcare educators. Prezio; 2014;[21]

US

RCT

Uninsured

Primary

Mexican

care

12

180

The

Community

Diabetes

Education (CoDE) program:  a one to one culturally tailored diabetes education

Americans with type 2 diabetes

32

Usual

At 12-month, the

care

intervention vs control:  more reduction in HbA1c (0.7%)

and management program  By community health workers. Mash;

South

2015; [22]

Africa

RCT

Subjects with

Primary

type 2 diabetes

care

12

1570

Group diabetes education

Usual

At 12-month, the

program:

care

intervention vs control:  more reduction in systolic blood

 4 sessions each lasting up to 60 min, focusing on what

pressure( 4.65 mmHg)

is diabetes, lifestyle

and diastolic blood

modification, understanding

pressure ( 3.3mmHg).

the medication and avoiding complications  By health promoters. Adibe 2013;[18]

Nigeria

RCT

Subjects

with

Tertiary

type 2 diabetes;

hospital

12

220

Pharmaceutical

care

Usual

At 12-momth, the

intervention (PC):

care

intervention vs control:

 monthly education program

offered

 incremental QALY gain

consisted of 4 sections of

by the

90-120 minutes, covering

hospital

diabetes overview and its complications, self-monitoring blood glucose techniques and interpretation of diabetes-related tests, medication, lifestyle modification, counselling

33

of 0.12 per subject

and effective interaction with health providers.  By pharmacist. Dijkstra;

Netherla

2005; [12]

nds

RCT

Subjects with

Secondary

type 2 diabetes

care

12

764

in outpatient hospital

Professional-directed group

Usual

At 12-month:

 education meetings for physicians and diabetes

hospital

 HbA1c decreased by 0.3% in the

specialist nurses by an

nt care

outpatie

patient-centred group

opinion-leader to discuss the

and by 0.1% in the

current guidelines.

professional-directed group but increased by

Patient-centred group:

0.2% in the control

 Patients were given a

group.

‘diabetes passport’ which outlined the national guidelines and aimed to educate; Education meetings were organized for patients with diabetes; Information leaflets and waiting room posters were also designed for patients.  not clear on the type of educator. Mason 2006;[13]

UK

RCT

Subjects from a

Primary

population-bas

care

12

591

Pro-Active

Call

Centre

Treatment Support (PACCTS):  sought to improve glycemic

ed register

34

Usual

At 12-month, the

care

intervention vs control:  more reduction in

control through patient

HbA1c (0.31%)

education about lifestyle and medication adherence, metabolic management and medication adjustment  By telecarer. Lewis 2014; [20]

US

Control

Data from five

led trial

1827

Five Alliance programs:

Subjects

Program group vs

alliance

particip



each focus on three core

not

comparison group:

programs

ants,

component including

participa



586

patient change, clinician

ted in

1827

without

change and system

the

subjects

particip

change.

alliance program

consistent

participated

of

in

Not clear

Not clear

ants



the

programs

Patient education included community, small group,

and

a

and individual materials as

(compar

comparison

well as classes and

ison

cohort of 586

discussions focused on

group)

without

evidence-based topics

participation.

related to diabetes self-management program 

35

Not clear on the type of educator.

s

more reduction in HbA1c (0.31%)

Table 3. Summary of costs and cost-effectiveness Author;

Perspecti

Year;

ve

Costing of Program Cost of program

Cost of health

Subject’

Subject’

Subject’s other

Total

service

s time

s travel

out of pocket

program

utilization

cost

cost

payment

cost/subje

ICER

ct Ritzwolle

Provider

Yes

r;

or payer

 staff cost

--

--

--

US$1295

--

usual care:

 equipment

2006;[23]

Mediterranean Lifestyle Program vs

 materials

 US$3,808 per incremental reduction in HbA1c (%),

 rental cost

 US$2,354 per unit reduction in body mass index (kg/m2),

 meals in the program

 US$196 per-gram reduction in intake of saturated fatty acids,  US$644 per unit improvement self-care summary score.

Ritzwolle

Health

Yes

Yes

r 2011;

plan and

 staff cost

 travel

[16]

participant

 equipment

 attend

 materials  overhead  other operating expenses

class

Yes

Yes

US$4634

 health club and gym

--

Viva Bien Program vs usual care:  US$7,723 per incremental reduction in HbA1c (%),

memberships  exercise equipment  changes in the cost of food due to

36

 US$7,723 per incremental reduction in BMI (kg/m2).

the program Schechte

Provider

Yes

--

--

--

US$176.6

--

Behavioural telephonic intervention

r;

 staff cost

vs received only the diabetes

2012;[17]

 tel. charges

self-management print materials:  US$490.6 per incremental reduction in HbA1c (%),  US$2617.4 per person achieving HbA1c<7%.

Hendrie;

Health

Yes

2014;

sector

 staff cost

Program vs standard pharmacy care:

 equipment

 US$39 per day of glycaemic

[19]

--

--

--

US$356

--

 materials

Diabetes Management Education

symptoms avoided

 tel. charges  flat fees to pharmacies Handley

Health

Yes

--

--

--

US$782

--

Automated telephone

2008;[14]

systems

 Start-up (e.g. setting up system,

self-management (ATSM) support

staff training,

usual care:

personnel time for

 US$65,167 per QALY gained for start-up and ongoing

with nurse care management vs

developing message and

implementation costs combined  US$32,333 per QALY gained for

protocols)  On-going (e.g. active nurse care

ongoing implementation costs alone.

37

management activities, patient recruitment and retention, monthly ATDM service costs) Gillett

NHS and

Yes

2010;[15]

personal

--

--

--

£203 based

Yes

Diabetes education and self

 staff cost

on trial

 drugs

management program vs usual care:

social

 materials

data;

services

 venue

£76 based

 GP and other primary care

 £5,387 per QALY gained based on trial data,

 staff travel cost

on real

consultation

 sundries

world.

 cost relates to

 £2,092 per QALY gained based on the “real world” intervention cost.

 refreshments

long-term complications

Prezio

Health

Yes

2014;[21]

system

Mash

Societal

US$0.68

Yes

The Community Diabetes Education

 staff cost

per day per

(not clear on the

program vs usual care:

 supplies used in the sessions

subject

component)

 US$355 per QALY gained over 20 years

US$20

Yes

Group diabetes education program

Yes

2015;

 staff cost

[22]

 materials

Yes

Yes

--

Yes

--

--

vs usual care:  healthcare costs related to  US$1,862 per QALY gained, cardiovascular based on the assumption of

 staff travel cost  SMS reminders

diseases

annual intervention cost and persistent effect

38

Adibe

Patient

--

--

--

Yes

US$159

 Self-monitorin g resources

2013;[18]

 Education

Yes

Pharmaceutical care intervention vs

 drugs

usual care:

 hospital care visit

 US$571 per QALY gained

 primary care visit

materials

 auxiliary health care  laboratory cost Dijkstra;

Not stated

Yes

--

--

--

£2 in the

Yes

professional-directed

 diabetes control

control:

 treatment for complications

patient-directed group versus control:

2005;

 staff cost

professiona

[12]

 materials

l-directed group;

group

 €32,218 per QALY gained,

patient-cen

 €16,353 per QALY gained, patient-centred group versus

tred group;

professional-directed group:

£3.5 in the

vs

 €881/QALY gained. Mason 2006; [13]

Not stated

Yes

--

--

--

£1,088

--

Pro-Active Call Centre Treatment

 staff cost

based on

Support vs usual care:

 venue rental

trial and

 £43,400/QALY under trial

 computer

best-guess

terminal rental

assumption

 commissioning costs

;

 running expenses

on routine

£714 based use and

39

condition,  £33,700/QALY under routine-use with a full caseload of patients with HbA1c≥7%.

best-guess assumption . Lewis

Health

Yes

2014;

care

[20]

system

--

--

--

US$975 in

Yes

Alliance

(not clear on the

the first

(not clear on the

group:

component)

year,

component)

US$61,011 per QALY gained under

“—“ means not included.

40

program vs comparison

US$520 in

the conservative scenario (program

the

reduced HbA1c by 0.31%, and has

subsequent

insignificant effect on blood pressure

year.

and total cholesterol).

Systematic review on the cost-effectiveness of self-management education programme

Table 4 Summary of the modelling Author;

Model type

simulated disease

Year

time

length of

horizon

program effect assumed

Gillett

Sheffield type 2

Estimating the effect of

2010;[15]

diabetes model which

alternative interventions on

is an integrated

long term incidence of

individual level

diabetic complications (i.e.

simulation model.

retinopathy, nephropathy,

lifetime

3 years

lifetime

Test:

neuropathy, coronary heart disease, cerebrovascular disease) and mortality, and the associated economic effects of such interventions and outcomes using changes in key risk factors, e.g. smoking status, HbA1c level, lipid concentration and blood pressure. Mash

Markov

Incorporated

2015;

micro-simulation model

cardiovascular risk factors,

[22]

developed using South

i.e.

African datasets.

diabetes, blood pressure,

age,

basic

sex,

 Persistent effect

smoking,

 one year effect

total cholesterol or body

 three year declining

mass index to predict future rates of angina, myocardial infarction,

stroke,

and

effect

death. Prezio

A validated Archimedes

Simulated

microvascular

2014;[21]

Model which was an

(retinopathy,

individual-level

and

simulation.

macrovascular

nephropathy

neuropathy)

(cardiovascular

and

20

not mentioned

years

and

cerebrovascular disease) Dijkstra

A probabilistic Markov

The relationship between

2005;

model that described

the HbA1c level and

[12]

the epidemiology of

progression towards

Dutch Type 2 diabetic

diabetic complications was

patients. It accounts for

from researchers’ earlier

ageing of patients, the

publication.

increase in HbA1c and the age-related 41

lifetime

not mentioned

Systematic review on the cost-effectiveness of self-management education programme

increase in the risk of complications. Lewis

The validated Centres

Simulated

2014;

for Disease Control and

neuropathy,

[20]

Prevention-RTI

coronary heart disease and

diabetes

stroke.

cost-effectiveness model

42

nephropathy, retinopathy,

lifetime

persistent effect

Systematic review on the cost-effectiveness of self-management education programme

Highlights  A small number of cost-effectiveness studies were identified with only eight of sufficiently good quality.  The cost of achieving reduction in clinical risk factors appears to be modest.  Self-management education programmes are likely to be cost-effective in the long-term.

43

Systematic review on the cost-effectiveness of self-management education programme

Appendix 1. Search terms in each database PubMed ((((“economic evaluation” OR “cost effectiveness” OR “cost consequence” OR “cost utility” OR “cost benefit” OR (“technology assessment" AND "biomedical”) OR “health technology assessment” OR “quality adjusted life year” OR “qaly”))) AND ((“type II diabetes” OR “non insulin dependent diabetes” OR (“diabetes mellitus" AND "type 2”) OR “NIDDM” OR “T2DM” OR “type 2 diabetes”))) AND ((“empowerment” OR “education” OR “health education” OR “knowledge improvement” OR (behavio* AND “change”) OR “lifestyle modification” OR “self management” OR “self care” OR (behavio* AND “therapy”) OR “psychotherapy” OR “self help group” OR “DSME” OR “DSMT” OR “DSMI” OR “DSMA”)) Cochrane 1. MeSH term Diabetes Mellitus, Type 2 explode all trees 2. ((type 2 or type ii or type two) NEAR/5 diabetes) 3. NIDDM 4. T2DM 5. “Non insulin dependent diabetes” 6. (#1 OR #2 OR #3 OR #4 OR #5) 7. MeSH term cost benefit analysis explode all trees 8. “economic evaluation” 9. “Cost effectiveness” 10. “Cost consequence” 11. “cost utility” 12. “technology assessment” NEAR/3 biomedical 13. “Health technology assessment” 14. “Quality adjusted life year” 15. “QALY” 16. (#7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15) 17. MeSH term Patient Education explode all trees 18. MeSH term Models, Educational explode all trees 19. MeSH term Self Care explode all trees 20. MeSH term Self Efficacy explode all trees 21. ((educat* or train* or learn* or teach* or empower* or behavior* or 44

Systematic review on the cost-effectiveness of self-management education programme

knowledge or manag* or chang*) NEAR/3 (patient* or self* or program* or model* or modif* or system* or therapy or interven*)) 22. Education 23. “Health Education” 24. “Knowledge improvement” 25. Behavio* NEAR/2 change 26. “Lifestyle modification” 27. “Self management” 28. “Self care” 29. Behavio* NEAR/2 therapy 30. Psychotherapy 31. “Self help group” 32. DSME 33. DSMT 34. DSMI 35. DSMA 36. (#17 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35) 37. (#6 AND #16 AND #36) PsycINFO (“economic evaluation” OR “cost effectiveness” OR “cost consequence” OR “cost utility” OR “cost benefit” OR (“technology assessment" NEAR/2 "biomedical”) OR “health technology assessment” OR “quality adjusted life year” OR “qaly”) AND (“type II diabetes” OR “non insulin dependent diabetes” OR (“diabetes mellitus" NEAR/2 "type 2”) OR “NIDDM” OR “T2DM” OR “type 2 diabetes”) AND (“empowerment” OR “education” OR “health education” OR “knowledge improvement” OR (behavio* NEAR/2 “change”) OR “lifestyle modification” OR “self management” OR “self care” OR (behavio* NEAR/2 “therapy”) OR “psychotherapy” OR “self help group” OR “DSME” OR “DSMT” OR “DSMI” OR “DSMA”)

45