Ebph.it

epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS A systematic review of the cost-effectiveness of lifestyle modification as primary prevention intervention for diabetes mellitus type 2 Katrin I. Radl(1), Carolina Ianuale(2), Stefania Boccia(2) Background: diabetes is one of the leading causes of death, and has a huge economic impact on
the burden of society. Lifestyle interventions such as diet, physical activity and weight reducing are
proven to be effective in the prevention of diabetes. To encourage policy actions, data on the cost-
effectiveness of such strategies of prevention programmes are needed.
MeThods: a systematic review of the literature on the cost-effectiveness of prevention strategies
focusing on lifestyle interventions for diabetes type 2 patients. a weighted version of drummond
checklist was used to further assess the quality of the included studies.
resuLTs: six studies met the inclusion criteria and were therefore considered in this paper. Intensive
lifestyle intervention to prevent diabetes type 2 is cost-effective in comparison to other interventions.
all studies were judged of medium-to-high quality.
concLusIons: policy makers should consider the adoption of a prevention strategy focusing on
intensive lifestyle changes because they are proven to be either cost-saving or cost-effective.
Key words: Diabetes; Prevention; Cost-effectiveness; Lifestyle (1) SDA Bocconi School of Management, Milan, Italy Corresponding author: Katrin Radl, Gletscherblick
(2) Institute of Hygiene, Università Cattolica del Sacro 30b, 6080 Igls, Austria. Tel: 0043 676 3302 448. Cuore, Rome, Italy the literature shows strong evidence that evitable risk factors like overweight and lack Diabetes mellitus is one of the top ten of physical activity are the main determining, causes of death for both low and high- non-genetic factors of DMT2 [3-7]. The income countries [1]. An interaction of genetic worldwide increase of the prevalence of predisposition along with behavioural and diabetes mellitus type 2, and the importance environmental risk factor was found to be a of obesity and lack of physical activity, as reason for diabetes mellitus type 2 (DMT2) stated by the WHO [8] increase the need of [2]. Even though it is not yet known what prevention strategies and policy implications role genetics play in occurrence of diabetes, to be taken. Caring for diabetes and the coSt-effectiveneSS of lifeStyle modification for dmt2 epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS subsequent complications is expensive [9], Quality Adjusted Life Years (QALYs) gained however some interventions have the ability or as Life Years Gained (LYGs); original to reduce the burden of the disease. One publication language was English; published possible primary prevention strategy relies between January 2007 and February 2012; on lifestyle modification i.e., changed dietary diabetes type 2. Studies were excluded if habits, increased physical activity, maintaining the lifestyle intervention was combined with or reducing body weight. In the last years therapy, or if they were compared with several randomised, controlled clinical trials treatments such as gastric bypass surgery or have investigated the impact of changing diet therapeutic or enteral nutrition.
and physical activity on the prevention of type 2 diabetes. The Finnish Diabetes Prevention Study (DPS) as well as many other studies reporting of results
[10-12], demonstrated a clinically significant impact of lifestyle changes in the decrease of In order to make ICERs (Incremental diabetes. However, resources are scarce and it Cost-Effectiveness Ratio) within the different is important to review if the adoption of such studies comparable, all costs were converted a prevention strategy is cost-effective. The aim into 2012 Euro using the OANDA Currency of this study is to systematically review the Exchange Calculator [14] for yearly average existing English literature available from 2007 annual exchange rates. Costs already expressed to 2012 on the cost-effectiveness of lifestyle in Euro (€) were also adjusted to 2012 by using changes as a primary prevention intervention the Italian Consumer Price Index [15]. ICERs for diabetes mellitus.
are then presented in €/QALY or €/LYG and rounded to the nearest hundred Euro.
For the classification of the intervention a research desIgn and MeThods
threshold approach was used, assuming that an intervention is cost-effective if the ICER is This systematic review was performed lower than € 20 000-30 000 and cost saving if following the PRISMA statement [13]. Studies the costs are lower and health outcomes are were identified by searching NCBI's PubMed better than in the compared intervention.
database and SCOPUS database. Following Results from the included studies are keywords were used to search the database summarised giving an overview of the study for appropriate literature indicating diabetes: population, the intervention, the comparison, diabetes OR diabetes mellitus OR diabetes where effectiveness data was drawn from, the AND mellitus OR diabetes insipidus OR analytical horizon, the study perspective and diabetes AND insipidus; indicating cost- the classification of the intervention using the effectiveness: cost-benefit analysis OR cost- criteria described earlier.
benefit AND analysis OR cost-effectiveness OR cost AND effectiveness; indicating prevention: prevention AND control OR prevention OR prevention and control OR prevention; indicating lifestyle: life AND style OR lifestyle A weighted version of Drummond checklist OR life style.
was used to further evaluate the quality of the studies included in the systematic review [16, 17]. The checklist was developed to assess the Inclusion and exclusion criteria
quality of an economic evaluation considering the following sections: study design, data Criteria for inclusion in the review were the collection, analysis and interpretation of results. following: original cost-effectiveness analysis All of the 35 items were explored by two or other full economic analysis; primary independent reviewers for included study. The intervention by changing lifestyle (e.g. by diet weighted version assigned a maximum global and/or physical activity); prevention strategy score of 26 for study design, of 45 for data for diabetes type 2 patients; pharmacotherapy collection, of 48 for analysis and interpretation or no intervention was a comparator to lifestyle of results section, while the global highest intervention; outcomes were presented as available score was 119.
coSt-effectiveneSS of lifeStyle modification for dmt2 epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS Study (DPS) [20]. By developing a Markov model with seven states over a lifetime horizon Following this search strategy, 128 studies they estimated that prevention strategies based were found in the PubMed database and 139 on lifestyle interventions could be cost-effective in SCOPUS database. All studies were then from the societal perspective but depended exported into EndNote. After exclusion of on sex, age group and threshold values. papers published before 2007, 145 abstracts However the results of the two studies were remained for screening. Six studies were quite different. The study result of 2007 [18] identified as being eligible for this review was that in borderline groups the cost/QALY meeting the inclusion criteria. Most of the was € 47/QALY for females and € 228/QALY papers were excluded because they were either for males. In the study conducted only one not related to diabetes or did not provide the year later [19], the ICERs were much higher, right outcome data (€/QALY or €/LYG) (Figure coming up to € 3 200/QALY for females and 1 for the Flow Diagram). A summary of the € 1 600/QALY for males at the age of 30 with studies included in this systematic review can overweight. The difference might result from be found in Table 1.
the different population and age considered in Two studies by Galani et al. [18, 19] focusing the result, but this still has to be considered on lifestyle intervention were conducted on when interpreting the results.
both obese and overweighed population Icks et al. [21] investigated the cost- groups in Switzerland. Effectiveness data was effectiveness of the Diabetes Prevention taken from the Finnish Diabetes Prevention Program (DPP) in a real-world setting in ThE sEaRch sTRaTEGy and Flow dIaGRam FoR daTabasEs sEaRch coSt-effectiveneSS of lifeStyle modification for dmt2 epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS Germany, getting their effectiveness data patient participation to achieve better clinical both from the DPS and literature review. and cost-effectiveness, because acceptance of They conclude that even though the lifestyle and adherence to the intervention is low intervention program was more cost effective and the drop- out rate is high. ICERs were than metformin, before implementing such a calculated from the societal and healthcare strategy efforts should be made to improve perspective and were reported as € 31 000/ dEscRIpTIon oF ThE cosT-EFFEcTIvEnEss sTUdIEs FoR dIabETEs pREvEnTIon yEaR, InTERvEnTIon compaRIson modEl classIFIcaTIon dIscoUnT analysIs €200/QALY 2006, 3% Probabilistic DAM Cost-effective €1 600/QALY 2006, 3% Probabilistic DAM Cost-effective years old and overweighted [18], 2007, intervention €10 900/LYG; 2000; 3% Univariate DAM Cost-effective [20], 2012, intervention 2012; 3% Probabilistic older age groups Probabilistic Markov coSt-effectiveneSS of lifeStyle modification for dmt2 epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS QUalITy oF InclUdEd sTUdIEs REFEREE's chEcklIsT (1) The research question is stated (2) The economic importance of the research question is stated (3) The viewpoint(s) of the analysis are clearly stated and justified (4) The rationale for choosing the alternative programmes or interventions compared is stated (5) The alternatives being compared are clearly (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 (8) The source(s) of effectiveness estimates used (9) Details of the design and results of effectiveness study are given (if based on a single study)(10) Details of the method of synthesis or meta- analysis of estimates are given (overview) (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 whom 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 adjustments for inflation or currency conversion are given (20) Details of any model used are given (21) The choice of model used and the key parameters on which it is based are justified (22) Time horizon of costs and benefits is stated (23) The discount rate(s) is stated (24) The choice of rate(s) is justified (25) An explanation is given if costs 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 (31) Incremental analysis is reported (32) Major outcomes are presented 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=not; NC=not clear; NA=not appropriate coSt-effectiveneSS of lifeStyle modification for dmt2 epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS QALY and € 5 400/QALY respectively.
(item 27), the choice of variables for sensitivity In order to assess the economic analysis is justified (item 28); moreover all the consequences of a DPS intervention in Sweden, studies compared relevant alternatives (item Lindgren et al. [22] created a simulation model, 30), reported incremental analysis (item 31), using effectiveness data from the DPS and cost answer to the study questions (item 33) and data from Swedish studies. Besides showing conclusions (item 34). Most studies stated that such a program would be both cost-saving research question (item 1) and the viewpoint of from the healthcare perspective and cost- analysis (item 3), justified the form of economic effective form a societal perspective, they also evaluation (item 7), stated the primary outcome prognosticated an increase of survival and the measure of economic evaluation (item 11), cost associated with it. ICERs were € 10 900/ described methods for quantities and estimation LYG and € 156 000/LYG, respectively.
of unit costs (item 17) and details of any model A four-state Markov model was used by used (item 20), justified the choice of model Neumann et al. [23] with data on transition used (item 21), stated the range over which probabilities from best available evidence, variables are varied (item 29), gave appropriate calculating costs from the perspective of society. conclusions (item 35). Details of the study Their results show that diabetes prevention design were not clearly given by Neumann intervention can be cost-effective, but the outcome and al. [23], Galani and al. [18] and Galani is very uncertain and depending on different and al. [19]. Finally some deficiencies have factors, such as age. QALYs gained through been highlighted referring to analysis and this strategy were low, but the ICER showed interpretation of results section. According to cost-effectiveness if initiation of the intervention the weighted Drummond's scale, the median was at an age of 70 years. For men and women quality score of selected studies was 93, with between 30 and 50 years the adoption of the a minimum score of 77, and a maximum score intervention leads to a saving in costs.
of 105 (Table 2). In study design question two In order to assess the cost effectiveness studies attained the maximum available score, of the DPP, the Diabetes Prevention Program no maximum available score was attained Research Group [24] performed an economic in data collection section and analysis and evaluation both from the health care and societal interpretation of results.
perspective using prospectively collected data on resource utilisation, cost and quality of life. Throughout their study they approved the cost-effectiveness of lifestyle intervention over metformin and the cost-saving compared Several studies such as the US DPP [25], to placebo. They reported € 19 400/QALY for the Finnish DPS [11] and the Indian DPP [26] females and € 27 600/QALY for males for a have already presented the efficacy of lifestyle population aged 70 years.
modification. Our systematic review showed that lifestyle modification as a prevention strategy is not only efficient but it is also cost Quality of characteristics included in the study
effective and/or in some cases cost saving. If the and quality of the included studies
costs of such an intervention could be further reduced, cost-effectiveness would increase. A Table 2 reports qualitative evaluation possible first step in this direction could be the assigned to each included study, according changing of the setting where the intervention to the 35 items exploring study design, data was provided. But also the different methods collection and analysis and interpretation of available (information delivery about lifestyle results. All the studies clearly defined the modifications in groups or on a one-by-one economic importance of research question basis) should be evaluated and improved.
(item 2), the form of economic evaluation (item However our study has several limitations: 6), the sources of effectiveness estimates (item first, we included only English-language 8); in all the studies quantities of resources are publications; second, most of these studies rely reported separately from their unit costs (item on different methodologies, include varying 16), currency and prices data are reported (item types of costs, have different outcomes or 18), the approach to sensitivity analysis is given measure outcomes differently and have a related coSt-effectiveneSS of lifeStyle modification for dmt2 epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS baseline risk which makes the comparison when evaluating the cost-effectiveness.
difficult. Also the country setting can influence Furthermore it would be advantageous to the outcomes. Another issue that should be have information about the cost-effectiveness of considered is that different healthcare systems, changes in public policies or public insurance population groups and values, clinical practices, incentives to physician, may have an impact on Eventually the impact of multiple different interventions should be evaluated as well, We considered a threshold approach for the because in a real-world setting patients follow classification of the cost-effectiveness, however more than one intervention at the same time.
setting a threshold is controversial. Since the It is important to keep in mind that UK [27], the USA [28] and Australians [29] all cost-effectiveness should not replace decision use different thresholds, the WHO recommends makers, but should inform them to help them basing the decision on the country's GDP [30]. making their decision. Cost-effectiveness is An intervention is considered as cost-effective not the only aspect to consider when deciding if the costs per DALYs are 1.3 GDP per capita. whether or not to adapt an intervention, Five of the six studies used economic decision because CEAs (Cost-Effectiveness Acceptability) models to evaluate the costs, thus providing do not take into account the benefits, utility, information at a much lower level than trials willingness to pay (both from a personal in a real world setting. But even with the help and a societal perspective), or any social, of Decision Modelling not all costs, values and legal or ethical issues that may occur when criteria that are relevant for decision makers adapting the intervention. Although the six can be captured. Furthermore they are based studies included in the systematic review are on assumptions and may not represent the of good quality, some deficiencies have been reality due to oversimplification. However DAMs documented in all the sections. Future studies (Decision Analysis Model) have the ability to will need to take better account of the items help to inform policy makers. In the future related to study design, data collection and economic evaluation of diabetes intervention results from a methodological point of view. should address the impact in real-world settings, This is also necessary to perform comparable because non-compliance, drop-out rates or and scientifically based economic evaluations.
attrition rates are very often not considered references
[1] WHO-The top 10 causes of death; 2011. Available from:
[6] Chan JM, Rimm EB, Colditz GA, et al. Obesity, fat distribution, and weight gain as risk factors for clinical index.html [Accessed April 1, 2012] diabetes in men. Diabetes care 1994; 17(9): 961-9 [2] Neel JV. Diabetes mellitus: a thrifty genotype rendered [7] Dowse GK, Zimmet PZ, Gareeboo H, et al. Abdominal detrimental by progress? Am J Hum Gen 1962; 14(4): 353 obesity and physical inactivity as risk factors for NIDDM [3] Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of and impaired glucose tolerance in Indian, Creole, and obesity, diabetes, and obesity-related health risk factors, Chinese Mauritians. Diabetes care 1991; 14(4): 271-82 2001. JAMA 2003; 289(1): 76-9 [8] Diabetes - The World Health Organization. Available [4] Helmrich SP, Ragland DR, Leung RW, Paffenbarger Jr RS. Physical activity and reduced occurrence of non-insulin- dependent diabetes mellitus. New Engl J Med 1991; [9] Jonsson B. Revealing the cost of Type II diabetes in [5] Colditz GA, Willett WC, Rotnitzky A, Manson JAE. Europe. Diabetologia 2002; 45(7): 5-12 Weight gain as a risk factor for clinical diabetes mellitus [10] Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention in women. Ann Intern Med 1995; 122(7): 481-6 of type 2 diabetes mellitus by changes in lifestyle among coSt-effectiveneSS of lifeStyle modification for dmt2 epidemiology Biostatistics and public Health - 2013, volume 10, number 2 SyStematic reviewS and meta- and pooled analySeS subjects with impaired glucose tolerance. New Engl J Finnish diabetes prevention study (DPS). Diabetes care Med 2001; 344(18): 1343-50 2003; 26(12): 3230-6 [11] Uusitupa M, Louheranta A, Lindstrom J, et al. The [21] Icks A, Rathmann W, Haastert B, et al. Clinical and cost- Finnish diabetes prevention study. Brit J Nutr 2000; effectiveness of primary prevention of Type 2 diabetes in a real world routine healthcare setting: model based [12] Steyn NP, Lambert EV, Tabana H Conference on on the KORA Survey 2000. Diabetic Med 2007; 24(5): "Multidisciplinary approaches to nutritional problems". Symposium on "Diabetes and health". Nutrition [22] Lindgren P, Lindstrom J, Tuomilehto J, et al. Lifestyle interventions for the prevention of type 2 diabetes. Proc intervention to prevent diabetes in men and women Nutr Soc 2009 Feb; 68(1): 55-70 with impaired glucose tolerance is cost-effective. Int J [13] Liberati A, Douglas G. The PRISMA statement for Technol Assess 2007; 23(2): 177 reporting systematic reviews and meta-analysis of studies [23] Neumann A, Schwarz P, Lindholm L. Estimating the cost- that evaluate health care interventions: explanation effectiveness of lifestyle intervention programmes to and elaboration. Italian Journal of Public Health 2009; prevent diabetes based on an example from Germany: Markov modelling. Biomed Central Ldt. 2011 [14] OANDA - Historical Exchange Rates Calculator; 2012. [24] Research-Group DPP. The 10-year cost effectiveness of lifestyle intervention or metformin for Diabetes asp?PageID=366 [Accessed April 4, 2012] Prevention. American Diabetes Association 2012 [15] OECD - Main Economic Indicators. Available from: [25] Knowler W, Fowler S, Hamman R, et al. 10-year follow- up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study. Lancet 2009; database_data-00052-en [Accessed April 4, 2012] [16] Drummond MF, Jefferson TO. Guidelines for authors and [26] Ramachandran A, Snehalatha C, Mary S, et al. The Indian peer reviewers of economic submissions to the BMJ. Diabetes Prevention Programme shows that lifestyle The BMJ Economic Evaluation Working Party. BMJ 1996; modification and metformin prevent type 2 diabetes in 313(7052): 275-83 Asian Indian subjects with impaired glucose tolerance [17] La Torre G, Nicolotti N, de Waure C, Ricciardi W. (IDPP-1). Diabetologia 2006; 49(2): 289-97 Development of a weighted scale to assess the quality [27] McCabe C, Claxton K, Culyer AJ. The NICE cost- of cost-effectiveness studies and an application to the effectiveness threshold: what it is and what that means. economic evaluations of a nti-HPV tetravalent vaccine. Pharmacoeconomics 2008; 26(9): 733-44 Italian J Public Health 2011; 19: 103-11 [28] Weinstein MC. How much are Americans willing to pay [18] Galani C, Schneider H, Rutten FFH. Modelling the for a quality- adjusted life year? Med Care 2008; 46(4): 343 lifetime costs and health effects of lifestyle intervention [29] Moodie M, Haby M, Wake M, et al. Cost-effectiveness in the prevention and treatment of obesity in of a family-based GP-mediated intervention targeting Switzerland. Int J Public Health 2007; 52(6): 372-82 overweight and moderately obese children. Econ Hum [19] Galani C, Al M, Schneider H, Rutten FFH. Uncertainty in Biol 2008; 6(3): 363-76 decision- making: value of additional information in the [30] Murray C, Evans DB, Acharya A, Baltussen R. cost-effectiveness of lifestyle intervention in overweight Development of WHO guidelines on generalized cost- and obese people. Value Health 2008; 11(3): 424-34 effectiveness analysis. Health Econ 2000; 9(3): 235-51 [20] Lindstr¨om J, Louheranta A, Mannelin M, et al. The coSt-effectiveneSS of lifeStyle modification for dmt2

Source: http://ebph.it/article/viewFile/8846/8037