6. Multimodal Training Intervention

Iceland’s population is ageing, which poses several challenges. Since 1980, the proportion of the population aged 65 years and over grew from 10% to 14%, and is expected to increase to 24% by year 2050 (Statistics Iceland, 2019[1]; Statistics Iceland, 2019[2]).1 Consequently the country has seen a rise in the number of people living with chronic diseases, greater demand for labour-intensive long-term care, and a decline in the proportion of the working age population (OECD, 2019[3]).

In response to these challenges, the Icelandic private company, Janus Health Promotion, developed the Multimodal Training Intervention (MTI), which was created and designed as a continuation of the doctoral project Multimodal Training Intervention – An Approach to Successful Aging (Guðlaugsson, 2014[4]). MTI aims to improve the fitness of participants enabling them to participate for longer in everyday activities, live longer in their own home, work for longer in the labour market, and delay or prevent admission to a nursing home.

MTI is targeted at those aged 65 years and over who are in good health. That is, people who:

  • Live independently

  • Are able to travel to and from training and seminar groups as part of MTI

  • Receive at least 6 out of a total 12 points in the SPPB (Short Physical Performance Battery) test, which is used to assess lower extremity in older, non-disabled adults (a score of 12 indicates the patient does not have any lower mobility limitations) (Guralnik et al., 1994[5]).

The intervention lasts for 24 months with activities broken into four sequential steps (see Table 6.2). These steps include endurance training (ET) (e.g. walking, cycling) and resistance training (RT) for all major muscles groups under the guidance of a professional trainer, as well as lectures on health and nutrition-related topics led by a nutritional counsellor. Physical activity classes are hosted at local indoor fitness centres, which have the necessary gym equipment.2 The focus on RT aligns with physical activity guidelines proposed by the World Health Organization (WHO), who recommend that those aged 65+ engage in this form of activity two to three times a week or more to “enhance functional capacity and to prevent falls” (World Health Organization, 2020[6]).

A key component of MTI is the collection of participant data every six months over the two-year period. It is the responsibility of employees of Janus Health Promotion (the private company responsible for MTI) to collect patient measurements including anthropometric (e.g. body-mass index (BMI), blood pressure, fat and muscle mass) and several physical activity outcomes (e.g. walking speed and the SPPB). Employees of Janus Health Promotion receive support from specialised surveyors who are trained in taking measurements for older age groups.

MTI includes several digital components. First, participants can track their performance by logging their workouts and diet in a dedicated mobile app. Second, municipalities have access to an online dashboard which displays results from each round of participant measurements. And third, MTI administrators have created a website and Facebook group to provide participants with important administrative information as well as direct contact with professional trainers and nutrition counsellors.

The cost of delivering MTI over a two-year period is approximately EUR 2093 per person or EUR 87 per month. In addition to the costs of providing RT, ET and nutrition and health education sessions, this figure covers promotion, marketing, the digital tools that allow participants to track their performance and stay in contact with MTI administrators, as well as costs associating with taking patient measurements.

To date, around 1 000 people from Iceland have previously or are currently participating in MTI across five municipalities (see Box 6.1 for a description of participant characteristics in Iceland).3 At any one time, between 80-160 people are enrolled in each municipality. Over the course of two years, around 25% of participants will drop out. As part of the European Commission’s Joint Action on Chronic Diseases, MTI was transferred to regions in Spain (Aragón) and Lithuania (Klaipėda).

This section analyses MTI against the five criteria within OECD’s Best Practice Identification Framework – Effectiveness, Efficiency, Equity, Evidence-base and Extent of coverage (see Box 6.2 for a high-level assessment of MTI). Further details on the OECD Framework can be found in Annex A.

OECD’s Strategic Public Health Planning for non-communicable diseases (NCDs) microsimulation model (SPHeP-NCDs model) was used to estimate the health and economic impact of expanding MTI across Iceland. Details on the model are in Annex A, while the list of model assumptions and limitations specific to the MTI analysis are in 0 of this document.

This section presents results for Iceland followed by remaining OECD and non-OECD European countries.

Expanding MTI to the whole of Iceland is estimated to lead to 7.08 life years (LY) and 8.17 disability-adjusted life years (DALYs) gained per 100 000 people, on average, per year over the period 2021-50. These figures translate into a cumulative gain of 456 LYs gained and 534 DALYs by 2050 (Figure 6.1).

In gross terms, MTI is expected to have the greatest impact on musculoskeletal disorders (MSDs) and cardiovascular diseases (CVDs) (Figure 6.2). Between 2021 and 2050, the number of MSDs and CVD cases is estimated to fall by 182 and 172 cases, respectively. Other diseases affected include mental health, diabetes, dementia and several cancers.

In proportional terms, MTI has the largest impact on diabetes (Figure 6.3). The number of diabetes cases averted as a proportion of total new diabetes cases for those aged 65+ (i.e. the target population) is estimated at 0.93%. The proportion of other diseases averted is lower, ranging from 0.03% for mental health to 0.15% for CVDs.

Transferring MTI to all OECD and EU27 countries is estimated to result in 7.7 and 9.4 LYs gained per 100 000 people (ranging from 2.8 in Norway to 18.9 in Bulgaria), respectively, on average, per year between 2021-50 (Figure 6.4). Regarding DALYs, the figure is even higher at 9.1 for OECD and 10.8 for EU27 countries.

In gross terms, MTI would have the greatest impact on CVDs with the intervention estimated to reduce the number of cases by 0.77 and 0.33 million in OECD and EU27 countries, respectively, between 2021-50 (Figure 6.5). This is followed closely by MSDs with cases estimated to decline by over 1.1 million across all countries. In proportional terms, MTI would have the greatest impact on diabetes and related cancers, with 0.71% and 0.19% of new cases avoided amongst the 65+ population, respectively (Figure 6.6).

Similar to “Effectiveness”, this section presents results for Iceland followed by remaining OECD and non-OECD European countries.

By reducing rates of obesity, MTI can reduce health care costs. Over the modelled period of 2021-50, the OECD-SPHeP NCD model estimates MTI would lead to cumulative health expenditure savings of EUR 11.3 per person by 2050 (Figure 6.7) or EUR 0.57 per person, per year. Cost savings however are offset by intervention operating costs (see Table 6.3). This is common for obesity interventions with the exception of those that target large segments of the population, such as mass media campaigns, given costs are spread over a large number of people (OECD, 2019[9]).

Average annual health expenditure (HE) savings as a proportion of total HE is 0.026% for both OECD and EU27 countries (Figure 6.8). On a per capita basis, this translates into average annual savings of EUR 0.55 and EUR 0.51 for OECD and EU27 countries, respectively.

Table 6.3 provides information on intervention costs, total health expenditure savings and the cost per DALY gained in local currency for all OECD and non-OECD European countries. Results from the analysis show MTI leads to large health expenditure savings, however, these savings are offset by high intervention costs. The results are not surprising given the intensity (e.g. small fitness classes led by personal trainers) and therefore the relatively high cost of operating MTI.

Eligible individuals with a low socio-economic status (i.e. priority group) were considered in the design of the intervention. Specifically, by heavily subsidising participation fees.

The impact of MTI on different priority population groups – such as by socio-economic status and ethnicity – are not available. Therefore, at present, it is unclear if MTI reduces existing health inequalities.

Longitudinal panel from an intervention group only was used to model the effectiveness and efficiency of MTI. Participant data was collected at the start of the intervention and at 6, 12, 18 and 24 months. The data explicitly controlled for age, location and gender, in addition, the study controlled for time-invariant confounders by using a fixed-effects regression model (e.g. race). A large number of outcome indicators were measured, which relied upon internationally recognised data collection methods and tools (e.g. the SPPB, 6-minute walking test, BMI, waist circumference) (see 0 for a full list of indicators).

The Quality Assessment Tool for Quantitative Studies rates the study as “strong” in two areas – reducing selection bias and using reliable and validated data collection methods (Effective Public Health Practice Project, 1998[10]). Conversely, in line with many public health studies, neither researchers nor participants were blinded from the study, which is the key feature to rank a study in the highest quality group. In addition, the proportion of participants who had all measurements taken was less than 60% – reasons for dropout and characteristics of those who did not complete the full two years were not explored. Details of the assessment are in Table 6.4.

A previous evaluation by Guðlaugsson (2014[4]) used a randomised control, crossover design study to determine the impact of MTI over an 18-month period. Findings from the RCT were positive with the intervention group recording statistically significant improvements in physical performance using the SPPB score, the 8-foot up and go test (a test for dynamic balance) and the six-minute walking test. Anthropometric measurements also improved with BMI falling by 1.6 and 1.8 points for men and women, respectively. Conversely, only men in the control group recorded a statistically significant increase in their SPPB score.

MTI operates in five of the 72 municipalities in Iceland and has enrolled around 1 000 people (mixture of participants who have completed, in the middle or at the start of the programme).

MTI places are capped in each municipality therefore the real participation rate cannot be calculated as there aren’t enough places for all eligible people who want to participate.

Dropout rates from MTI are recorded every six months. Data from Janus Health Promotion in years 2017-18 indicate dropout is consistent over the 24-month period:

  • 0-6 months: 6.25% (from 160 to 150 participations)

  • 7-12 months: 6.67% (from 150 to 140 participants)

  • 13-18 months: 7.14% (from 140 to 130 participants)

  • 19-24 months: 7.69% (from 130 to 120 participants).

Participation and dropout rates by different population groups are not available. However, dropout shouldn’t necessarily be seen in negative terms, as participants may leave if they feel they have become self-sufficient, which is the ultimate objective of MTI.

This section summarises policy options available to policy makers and MTI administrators in settings where MTI is implemented (or being transferred) to further enhance the performance of this intervention.

MTI performs highly against the effectiveness criterion, therefore, no additional policies have been listed for MTI to enhance effectiveness.

Policies to enhance efficiency have not been identified for MTI.

A country health profile of Iceland in 2019 highlighted social inequalities within the population. For example, the gap in life expectancy at 30 between those with the highest and lowest level of education is 3.6 and 5 years for women and men, respectively. This gap is in part due to differences in risk factors such as obesity (OECD and WHO, 2019[11]).

Rising inequalities in health indicate MTI would benefit from focusing on recruiting participants from priority population groups. A study on effective strategies to recruit participants from lower socio-economic groups into a community-based lifestyle intervention included (Stuber et al., 2020[12]):

  • Multiple recruitment strategies to enhance familiarity with the intervention

  • Partnering with existing organisations that have close ties with the target group (e.g. social services and charities)

  • Involving trusted community members during the recruitment stage

  • Shortening the time period between recruitment and participation.

To extend the evidence-base of MTI, future evaluations would benefit from obtaining health care administrative data for participants and a comparable population group (from before participants receive MTI until it concludes and for a period thereafter). This information would allow researchers to develop a more in-depth understanding on the impact of MTI on disease incidence and health care costs, as well as analyse the impact on health care utilisation measures (e.g. hospital admissions for falls). More robust data and analysis will help secure political support to scale-up MTI across Iceland as well as transfer to other countries.

MTI administrators currently collect data for a wide range of internationally recognised indicators (see 0 for a full list of indicators). Consideration could be given to expanding data collection to include commonly used diet-related outcomes, including fruit and vegetable consumption (at least once a day or the national recommended level). Other commonly used indicators are listed below, however, these are more administratively burdensome to collect and may not be appropriate:

  • Sugar intake (less than 10% of total calorie intake)

  • Salt consumption (less than 5 grammes per day)

  • Saturated fatty acid intake (less than 10% of total calorie intake)

  • Average number of calories consumed per day

  • Wholegrain consumption in grammes per day.

In addition, where possible, it is recommended administrators collect data on a wider range of confounding variables including ethnicity, marital status and socio-economic status. This information could also be used to assess the impact of MTI across different priority population groups, and therefore determine whether MTI reduces existing health inequalities.

A significant barrier cited by MTI officials in Iceland is the difficulty in ensuring participants have their measurements taken every six months over a two-year period (Stegeman et al., 2020[13]). Using latest available data (2017-19), 43% of people who participated from the start to the end of MTI had all their measurements taken. This is problematic if measurement dropout (and therefore missing data) is not random as evaluation results may not reflect the participating population. For example, if participants with lower levels of motivation are less likely to be measured, the impact of MTI may be overestimated. Commonly employed techniques to reduce dropout between measurement rounds include:

  • Offering participants incentives for participating in measurements

  • Identifying characteristics of participants who are less likely to have their measurements taken and understand why and therefore potential solutions

  • Educating participants on the importance of having their data collected, not only at the individual level (to measure their own progress), but at the wider intervention level, for example, to secure future funding. For example, a seminar could be dedicated to teaching participants how to interpret their data and how it will be used and stored.

  • Providing participants with regular reminders in the lead up to measurements, and, if possible, increasing flexibility of when data can be collected (e.g. longer opening hours)

  • Minimising patient time spent having their measurements taken, including any waiting time

  • Focusing on individuals who experience greater difficulty attending measurement sessions such as those who do not live close by. For example, in Lithuania where MTI was transferred, administrators noted it was difficult to motivate participations from the rural Klaipėda district who had to travel long distances to attend monitoring activities (Stegeman et al., 2020[13]).

If significant levels of dropout between measurement periods continue, researchers could consider several techniques for dealing with missing data, other than complete case analysis (e.g. multiple imputation). The right technique will depend on the nature of the missing data.

Better knowledge on the benefits of exercising in old age may boost uptake in MTI. Feedback from Janus Health Promotion highlighted the important role stakeholders such as local government have in educating the older population on the health, social and economic benefits of exercising in older age – e.g. living independently for longer, working for longer and prolonging life. Improved health literacy may in turn boost motivation levels to exercise and therefore enrolment in MTI.

Finally, feedback from Janus Health Promotion highlighted the importance of educating the older population on the benefits of exercising.

This section explores the transferability of MTI and is broken into three components: 1) an examination of previous transfers; 2) a transferability assessment using publically available data; and 3) additional considerations for policy makers interested in transferring MTI.

As part of the Joint Action on Chronic Diseases (JA Chrodis Plus), MTI was transferred to two locations outside Iceland: Utebo in Aragón, Spain, and the Klaipėda district and city municipalities in Lithuania. In both locations, MTI was implemented in its original form.

To ensure a smooth transfer, representatives from Spain and Lithuania visited Iceland to learn “how everything works” such as taking patient measurements, running lectures, working with data and communicating with patients. A second visit was made by MTI good practice owners in Iceland to Spain and Lithuania to help them get started and to make any necessary corrections, particularly in regard to strength training requirements (Stegeman et al., 2020[13]). The two visits were seen as critical to a successful transfer.

Findings from JA Chrodis Plus evaluation reports indicate the transfer was successful, for example, in Lithuania, MTI improved participant physical activity levels, flexibility, endurance and vigour, which led the Ministry of Health to roll-out MTI across the country (Stegeman et al., 2020[13]). The success MTI in other locations is based on the same outcomes indicators used by MTI Iceland (see 0), such as the 6-minute walking test, blood pressure and SPPB scores. However, the two sites in Lithuania and Spain did not use all the questionnaire available to avoid over complication at the beginning.

The following section outlines the methodological framework to assess transferability and results from the assessment.

Details on the methodological framework to assess transferability can be found in Annex A.

Indicators from publically available datasets to assess the transferability of MTI are listed in Table 6.5. These cover indicators related to the population, political and economic contexts. Please note, the assessment is intentionally high level given the availability of public data covering OECD and non-OECD European countries.

Results from the transferability assessment are summarised below showing mixed results. For country level-results, see Table 6.6:

  • Across OECD and non-OECD European countries, a similar proportion of the population aged 65 years and over would be eligible for the intervention (as measured by the proportion of people accessing long-term and severe home-care services) indicating a similar extent of coverage.

  • MTI is likely to receive political support given most (90%) countries have an unhealthy eating and physical inactivity national action plan.

  • Higher levels of wealth in Iceland indicate users in other countries may experience greater financial barriers to accessing MTI if they incur out-of-pocket expenses – Iceland’s GNI (gross national income) per capita is USD 54 095 (PPP (purchasing power parity) international dollars) compared to an average of USD 42 103 in other OECD and non-OECD European countries).

  • Relatively low levels of internet users among the older population compared to Iceland may make it difficult to interact with participants and therefore keep motivation levels high over the two-year period.

Although not reported in the transferability assessment below, differences in population ethnicity may affect transferability. For example, the vast majority of citizens living in Iceland are white and born locally, which is different for example, from the United States, where over one in ten people are Black or African American (United States Census Bureau, 2019[19]).

To help consolidate findings from the transferability assessment above, countries have been clustered into one of three groups, based on indicators reported in Table 6.5. Countries in clusters with more positive values have the greatest transfer potential. For further details on the methodological approach used, please refer to Annex A.

Key findings from each of the clusters are below with further details in Figure 6.9 and Table 6.7:

  • Countries in cluster one, which includes Iceland, have political and economic arrangements in place to support MTI and therefore have conditions in place to readily transfer MTI to their local context.

  • Countries in cluster two, before transferring MTI, may want to undertake further analysis to assess whether the intervention is affordable among older populations who would need to pay out-of-pocket. It is important to note that Lithuania, which transferred MTI and recorded positive outcomes, falls under this cluster indicating the intervention can operate successfully in countries with lower levels of individual wealth.

  • Remaining countries are in cluster three, which may want to ensure MTI aligns with overarching political priorities before transferring the intervention.

Data from publically available datasets is not ideal to assess the transferability of MTI. For example, there is no international data on gym space and equipment, which is key to delivering MTI. Therefore, Box 6.3 outlines several new indicators policy makers should consider before transferring this intervention.

MTI has been shown to significantly improve a range of key health indicators including BMI, blood pressure and physical activity levels. Using OECD’s SPHeP-NCD model, it is estimated that scaling-up MTI across Iceland would lead to 7.08 life years (LY) and 8.17 disability-adjusted life years (DALYs) gained per 100 000 people, on average, per year over the period 2021-50. By reducing disease incidence across Iceland, MTI is expected to lead to cumulative health expenditure savings of EUR 11.3 per person by 2050. However, intervention costs outweigh health expenditure savings, which is common for obesity interventions.

As part of the European Commission’s JA Chrodis Plus, MTI has been successfully transferred to regions in two other European countries – Spain and Lithuania. Using publically available data to assess transferability to other OECD and non-OECD European countries, however, shows mixed results. For example, MTI is likely to receive political support as it addresses key government priorities (e.g. unhealthy eating), nevertheless, affordability may be an issue if participants are required to pay for MTI out-of-pocket. Based on feedback from owners of the intervention, there are two key factors behind the successful transfer of this intervention: first, a close relationship between the owner and transferring country, and second, implementation that is as close as possible to MTI’s original form.

Box 6.4 outlines next steps for policy makers and funding agencies regarding MTI.


[10] Effective Public Health Practice Project (1998), Quality assessment tool for quantitative studies, https://www.nccmt.ca/knowledge-repositories/search/14.

[15] Eurostat (2019), Self-reported use of home care services by sex, age and level of activity limitation.

[4] Guðlaugsson, J. (2014), Multimodal Training Intervention: An Approach to Successful Aging, University of Iceland, https://www.janusheilsuefling.is/wp-content/uploads/2019/06/Doktorsritger%C3%B0-Janusar-Gu%C3%B0laugssonar-12-9-14-III.pdf.

[7] Guðlaugsson, J., L. Janusdóttir and D. Janusson (2019), Multimodal Training Intervention in Municipalities: An approach to Successful Aging, Janus Health Promotion.

[5] Guralnik, J. et al. (1994), “A Short Physical Performance Battery Assessing Lower Extremity Function: Association With Self-Reported Disability and Prediction of Mortality and Nursing Home Admission”, Journal of Gerontology, Vol. 49/2, pp. M85-M94, https://doi.org/10.1093/geronj/49.2.m85.

[20] Hajek, A. and H. König (2018), “Are changes in body-mass-iandex associated with changes in depressive symptoms? Findings of a population-based longitudinal study among older Germans”, BMC Psychiatry, Vol. 18/1, https://doi.org/10.1186/s12888-018-1748-1.

[16] OECD (2020), C5B: Individuals using the Internet - last 3 m (%), OECD, Paris.

[3] OECD (2019), Health at a Glance 2019: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/4dd50c09-en.

[14] OECD (2019), Long-term recipients in institutions (other than hospitals) - total recipients over 65, % of total population aged 65+, OECD, Paris.

[9] OECD (2019), The Heavy Burden of Obesity: The Economics of Prevention, OECD Health Policy Studies, OECD Publishing, Paris, https://doi.org/10.1787/67450d67-en.

[11] OECD and WHO (2019), State of Health in EU - Iceland: Country Health Profiles 2019, European Commission, https://ec.europa.eu/health/sites/health/files/state/docs/2019_chp_is_english.pdf.

[8] Statistics Iceland (2020), Population by country of birth, sex and age 1 January 1998-2020.

[2] Statistics Iceland (2019), Population by sex and age 1841-2019.

[1] Statistics Iceland (2019), Population projection by age and sex 2019-2068.

[13] Stegeman, I. et al. (2020), D5.3 Recommendations for the implementation of health promotion good practices: Building on what works: transferring and implementing good practice to strengthen health promotion and disease prevention in Europe, EuroHealthNet & Finnish Institute for Health and Welfare.

[12] Stuber, J. et al. (2020), “Successfully Recruiting Adults with a Low Socioeconomic Position into Community-Based Lifestyle Programs: A Qualitative Study on Expert Opinions”, International Journal of Environmental Research and Public Health, Vol. 17/8, p. 2764, https://doi.org/10.3390/ijerph17082764.

[21] Torres-Reyna, O. (2010), Getting Started in Fixed/Random Effects Model using R, Princeton University, http://www.princeton.edu/~otorres/Panel101R.pdf.

[19] United States Census Bureau (2019), Population estimates.

[18] WHO (n.d.), Global Health Observatory, https://www.who.int/data/gho (accessed on 25 August 2021).

[17] World Bank (2017), GNI per capita, PPP (constant 2017 international $).

[6] World Health Organization (2020), Physical activity, https://www.who.int/news-room/fact-sheets/detail/physical-activity (accessed on 29 January 2021).

The effectiveness of MTI was first assessed using data provided by Janus Health Promotion. The data consisted of 57 variables (excluding self-reported survey results) which covered age, gender, a unique participant identifier, location as well as a range of outcome (e.g. BMI, systolic blood pressure and physical activity). Data for each of these variables was collected for 894 participants over five points in time covering two years (i.e. five measurement points collected six months apart, with the first measurement taken at the start of the intervention). On average, data for each participant was collected between 3-4 times.

To estimate the change in primary outcome measures (i.e. BMI, systolic blood pressure and physical activity4), a fixed effects regression method was used. A fixed effects regression was deemed suitable for the following reasons (Hajek and König, 2018[20]; Torres-Reyna, 2010[21]):

  1. 1. The data provided was longitudinal form using information from the same individuals

  2. 2. The dataset does not capture all relevant variables for explaining differences in risk factors (e.g. income, education, ethnicity, motivation levels and genetic disposition), that is, there are omitted variables

  3. 3. The omitted variables are unlikely to change over time, further, their effect on individuals is also time-invariant (i.e. the impact of income on BMI is the same at measurement 1 and measurement 5).

Fixed effects regressions were undertaken for each risk factor as the dependent variable – BMI, systolic blood pressure and MET min/week – with the following controls: age and measurement period. As outlined above, unobservable independent variables that are time-invariant are also taken into account with fixed effects, for example, as income, race and education). Regressions were run for men and women separately, therefore results also take into account MTI’s impact by gender.

Results from each fixed effects regression model found statistically significant improvements in all risk factors across all measurement periods. For example, BMI declined, on average, by 0.50 (p < 0.05) over the 24-month period for women and by 0.69 for men (p < 0.05).

The impact of MTI on BMI, systolic blood pressure and MET min / week was modelled as a linear increase/decrease over the 24-month period.

Please note, the analysis below is focused on BMI, systolic blood pressure and physical activity (in MET min/week) given these risk factors are present within OECD’s Strategic Public Health Planning for NCDs (SPHeP-NCD) model. Results for other outcome measures and previous MTI analyses are in 0.

There are limitations to the methodology used to model effectiveness. First, the change in outcome measures were calculated using measurements collected for participant data only (i.e. there was no control group). For this reason, the change in outcome variables caused by MTI may be over- or under-estimated. Second, survey weights were not available, therefore, changes in outcome variables may be over- or under-represent certain groups in the population. Third, not all participants had measurements taken and it is not clear whether these data are missing at random or not.

Based on feedback from Janus Health Promotion, the model assumes 60%5 of those who participate in the 24-month programme will participate in a second round. However, the intensity of each subsequent round will decline (e.g. lower number of training sessions and education seminars), thereby reducing impact on outcome indicators of interest. To take this into account, the model assumes for the second round of MTI, the proportion of services accessed declines by 42%6 with a proportional reduction in the impact on outcome measures (e.g. BMI). Users can participate in a maximum of two rounds of MTI.

The target age is men and women aged 65 years and over.

Exposure was calculated using the following three inputs:

  • 26% of the target age apply to be part of MTI (given the number of places in MTI were capped, the actual figure from MTI could not be calculated, therefore, the figure used represents findings from a previous analysis regarding update of prescription physical activity programs) (see Table 6.1 in OECD’s Heavy Burden of Obesity report (2019[9])).

  • 96% of those who apply are considered eligible to participate (figure provided by Janus Health Promotion)7

  • 99% who apply and are eligible end up participating in MTI (figure provided by Janus Health Promotion)

  • Overall participation rate = 0.26 * 0.96 * 0.99 = 24.7%

Dropout rates were provided by Janus Health Promotion.

The cost per participant per year is ISK 166 800 (using data provided by Janus Health Promotion). The costs of the programme cover:

  • Promotion and presentation of MTI: promoting MTI to eligible participants and other interested parties.

  • Website and Facebook group: providing regular updates regard the progress of the project, and connecting participants and trainers and each other.

  • Measurement collection: collecting data at baseline and throughout the intervention (five different time points).

  • Training costs: organised training sessions under the supervision of one or more trainers.

  • Education lectures: between 6-10 live educations run per year which include experts in various fields.

  • Mobile app: app for participants which allows them to log their workouts, view workout plans as well as populate their food diary and other statistics.

  • Online dashboard: municipalities have access to an online dashboard to view MTI’s progress – e.g. measurement results.

  • Information updates: detailed updates provided to municipalities every six months within a report format.

  • Meetings and “other”: meetings and other items that occur during the intervention.

The impact of MTI on a selection of other outcome measures of interest are listed in Annex Table 6.B.1 (for men and women combined). Outcome measures include: resting heart rate (heart beat per minute); waist-to-hip ratio (cm/m); 30-second chair stand (number of repetitions); short physical performance battery (total score); hand grip strength (both hands); six minute walking test (in metres); life quality (EQ-5D-5L scores); muscle mass (kg); body fat mass (kg); and body fat percentage (%). Overall the results are positive and statistically significant.

The impact of the multi-modal (MTI) training programme on key physical and mental outcome measures were first analysed and reported in a study by Guðlaugsson (2014[4]). Specifically, the study reported outcome measures for two groups, by gender, over an 18-month period:

  • Group 1 (immediate): accessed the MTI programme for the first six months only

  • Group 2 (delayed): accessed the MTI programme during the six and 12-month period only.

Changes in outcome measures for groups 1 and 2 (G1 and G2, respectively) are summarised in Annex Table 6.B.2 with results presented at percentage changes. Key findings include:

  • a statistically significant reduction in BMI for men and women in both groups

  • a statistically significant improvement in the 8 foot up-and-go test for men and women in both groups

  • a statistically significant increase in physical activity for G2 only, particularly for women.

Measurements were also recorded at the 18-month period with results for G1 and G2 combined (i.e. baseline to 18-months). The results indicate the MTI programme has a positive, lasting impact for men and women with statistically significant improvements in several outcomes measures (e.g. BMI and the short physical performance battery test) (Guðlaugsson, 2014[4]).

Changes in key outcome measures are also available for years 2017-19, however, results have not been tested for statistical significant (Annex Table 6.B.3). Between June 2017 and June 2019, MTI participants experienced improvements in all outcome measures, in particular, the number of strength training sessions per week (+83%) and daily activity (e.g. walking) (+58%) (Guðlaugsson, Janusdóttir and Janusson, 2019[7]).


← 1. Population projections are based on the “medium” trajectory estimated by Statistics Iceland (Statistics Iceland, 2019[1]).

← 2. Types of gym equipment include treadmills, stationary bikes, rowing machines, and weight machines.

← 3. Three new municipalities in Iceland will join MTI once COVID-19 restrictions have been lifted.

← 4. Physical activity was transformed into MET min / week based on the assumption that recorded activity represented “moderate activity” (50-60% of max heart rate) and therefore equivalent to a MET of 3.5.

← 5. Based on feedback from Janus Health Promotion, in Hafnarfjörður, approximately 70% of participants requested to participate in a second round compared to 50% in Reykjanesbaer,

← 6. This figure is based on feedback from Janus Health Promotion whereby the number of sessions declines by between 33-50% (average of 42%) for subsequent rounds.

← 7. This figure aligns with findings from OECD’s Measuring Social Care Protection for Long-Term Care in Old Age questionnaire for Iceland (2019), which found approximately 7% of those aged 65 years and over receive social home care and/or home nursing care services.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2022

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.