Effects of Primary Prevention in Elderly People-The Healthy Aging Initiative (HAI)

Effects of Primary Prevention on Non-communicable Disease in Elderly People-An Intervention Study in 70 Year Old Men and Women, the Healthy Aging Initiative

Hypothesis and aims The number of elderly will increase rapidly in Europe in the next 30 years, which affects the country's economic and social development, welfare, health care, and also the individuals. Future challenges include creating opportunities for people to maintain a high well-being in later stages of life, and coping with diseases and disability. Physical exercise and avoiding obesity have the potential to increase quality of life in the elderly by preventing CVD and stroke.

The proposed project is a intervention study where the overarching aim is to evaluate whether a primary prevention with the focus of decreasing obesity and increasing objective measures of physical activity will decrease the future risk of the endpoints cardiovascular disease, stroke, diabetes, falls, fractures, depression, dementia and death, in a population based cohort of 70-year-old women and men.

Specific aims: In a population based cohort of 5000, 70-year old women and men;

1. Investigate the association between risk factors assessed at baseline, with focus on objective measures of physical activity and body fat distribution, and the endpoints described above.
2. To investigate whether individuals given the prevention will have a lower future risk of the endpoints described above than 70-year olds in general Swedish population, after adjustments for differences in the different covariates at baseline.

In a second part, a follow up will be conducted after 5 years where all participants will be tested again. One aim is to evaluate whether the changes in the risk factors assessed at baseline, with focus of objective measures of physical activity and body fat distribution, are associated with the risk of non-communicable disease.

Working plan

Project overview- The Healthy Aging Initiative (HAI) HAI is a population-based health examination and intervention where all individuals living in Umeå municipal area are invited to participate when they are exactly 70 years old. Information from population registers is used to contact eligible individuals, who received printed information about the research project. Telephone contact is made shortly thereafter, during which individuals accept or decline the invitation to participate. As we aim to investigate a sample representing the general population, no eligible participant is excluded. The 70-year-old participants visit the clinic at 2 occasions a week apart. At the first visit, traditional and potential novel risk factors for non-communicable disease are measured (please see details below), blood samples are sent for analyses, and participants are sent home with accelerometers to collect data on physical activity and sedentary behavior. At the second visit, all participants are informed about the results of tests performed. They also receive counseling with focus on physical activity using motivational interviewing which is a directive patient-cantered counseling style to help the participants to explore and resolve their ambivalence about behavior change. Based on the results, the participants will also be given medical advice, or referred to primary care for follow up. They are also contacted at 6 months, 12 months per telephone for follow up, and we are planning for an in-clinic follow up after 5 years. The project started 5 years ago with the aim to include 2500 men and 2500 women. October 12, 2017, 3700 individuals have been included with a participation rate of 71%.

Questionnaires, basic measurements and blood samples All participants are asked to answer a comprehensive questionnaire organized in the following themes; socio-demography, lifestyle, mood, cognition, living habits, woman´s health, and medical history. Height, weight, forced expiratory flow, waist-hip ratio, blood pressure, blood glucose and blood lipids are measured/analyzed during the visit using standardized methods. In addition all subjects are asked to leave blood samples for later analyses of markers for primarily cardiovascular disease. All subjects participating in the project are also asked to leave blood samples for DNA extraction. In this DNA, candidate genes will be analyzed and finally a GWAS including 2.4 million genetic markers will be performed for the outcomes of physical activity, muscle strength and body fat distribution, blood pressure, blood glucose and blood lipids. Blood samples are stored in the local biobank until the preplanned analyses of risk factors for non-communicable disease presented above.

Measurements of body fat distribution and ectopic fat Peripheral quantitative computed tomography (pQCT 2000, Stratec, Germany) is used to obtain slices (taken as a percentage of limb length from the distal end of tibia, is used to measure bone density and muscle density, that is an estimate of ectopic fat accumulation at the 66% tibial scan site. Additionally, areal bone density, visceral fat, subcutaneous fat and gynoid fat are measured using dual-energy X-ray absorptiometry (DXA) and the new Corescan soft ware using a Lunar I-DXA (GE Healthcare, Wauwatosa, WI, USA).

Objective measures of physical activity Objective measures of physical activity is measured using a triaxial accelerometer (GT3X+, Maribo Medical, Denmark) during 1 week. The accelerometer is worn around the waist as this place has been shown to produce the most accurate results during validation.

Measurements of muscle strength, balance and gait Isometric muscle strength of the non-dominant hand is measured using a validated isokinetic hand dynamometer. Postural instability is studied when standing on a force plate with preferred placements of the feet. The instability is estimated by calculating the variability in the centre of pressure in relation to the supported foot area - with and without vision. The methods for testing have been developed exclusively for this project since clinical measurements used today are lacking in sensitivity. Details surrounding specific methods have been development in a joint cooperation with researchers from the Department of Integrative Medical Biology, Physiology unit (Professor Benoni Edin and Göran Westling, PhD).

Finally, we measure gait variability using a 6-meter gait test and the GAITRite system (CIR Systems, Sparta, NJ, USA). There are two separate tests. The first is conducted using self-selected speed without performing any tasks. The second is performed at a self-selected speed while performing a dual-task to increase the involvement of higher cortical functions on the gait variability. Balance and gait are of interest in the present project as measures of frailty, and of interest with respect to the risk of falls and fractures.

After the health investigation when all data has been collected All participants are informed about the results of the tests performed. Participants are also given advice based on their results with special emphasis on increasing the amount of physical activity. They are also advised to search medical care or referred based on risk factor evaluation.

Identification of non-communicable disease within the cohort. Information on diagnosis from 1987-01-01 and onwards is obtained by record linkage with the National Hospital Discharge Register (HDR), covering all public inpatient care in Sweden, and the National Hospital Outpatients Register, covering all public outpatient care in Sweden from 2003-01-01, administered the National Board of Health and Welfare in Sweden. Diagnosis were recorded using the International Classification of Diseases (ICD) version 9 (1987-1996) and version 10 (1997 and later). Deaths occurring during the study period are collected through the National Cause of Death Register, administered by the Center for Epidemiology at the National Board of Health and Welfare in Sweden.

Prescribed drugs for every subject in the cohorts based on social security number is available from the National Prescription Database at the National Board of Health and Welfare from 2005-07 and onward. Permission for merging data files has been obtained from the National Board of Health and Welfare. Socioeconomic data will be searched at Statistics Sweden (www.scb.se).

Statistical methods Differences between two different groups will be investigated using student's t-test for independent samples and differences between 3 groups will be investigated using Analysis of variance with Bonferroni's post-hoc test. Association between risk factors and future risk of CVD, stroke and diabetes will be investigated using multivariable adjusted Cox regression models. Cox regression models will also be used to investigate changes in the different risk factors during the 5 years of follow up in relation to the later risk of CVD, stroke and diabetes.

Statistical power To estimate the statistical power with respect to the effect of the cohort investigated, we first evaluated the risk of disease and death in the total Swedish population of 70-year old men and women that lived in Sweden 2006 (n=72375). During 5 years of follow up, 8969 subjects died, 2695 individuals suffered a first myocardial infarction and 2855 suffered a stroke, according to national registers. If we assume a similar risk of CVD, stroke and death in our cohort there will be 620 deaths and 383 incident cases of CVD or stroke during 5 years of follow up. Thus, also under two-tailed hypothesis testing, the total number of outcomes will enable detection of numerically small associations, and small differences in associations. Concerning the evaluation of the intervention given among the HAI participants, compared to the rest of the Swedish population of 70 year old men and women, we assume that the intervention will reduce the future risk of the combined endpoint death, myocardial infarction and stroke by at least 15%. If we then assume a sample size of 3500 individuals (in each group) and a significance level of 0.05, we will have 96% power to detect a difference. Given that the control group will be much larger, the statistical power is likely to be adequate also for detecting smaller differences between the groups.

Ethical considerations Only the research leader will have access to social security numbers from where the identity can be traced from individual data. No individual data will be presented. Permission for these studies has been obtained from the Regional Ethical Review Board in Umeå and conforms to the World Medical Association Declaration of Helsinki (Dnr 07-031M with extensions), and for the genetic part (06-134 and 09-029).

Significance and clinical relevance Non-communicable disease is the main cause of death and morbidity worldwide. The number of elderly will increase rapidly in the next 30 years and currently there are severe economic problems for Swedish health care with decreasing number of hospital beds and personnel as a result. The idea is that the very old should live at home healthy longer and that health care should be provided outside hospitals and specialized care. In another project we have investigated the incidence of myocardial infarction and stroke in the total Swedish population above 50 years of age, i.e. in 3.29 million individuals. In this cohort, 28,268 individuals suffered a myocardial infarction and 29,708 individuals suffered a stroke at a mean age of about 77 years in 2006. Thus, any preventive measures that could decrease the future risk of CVD and stroke in our cohort of 70-year olds would be of high importance. Furthermore, the potential of preventive measures in our cohort is high given that; 41% of the participants collected so far have systolic blood pressure above 140, 16% are obese (BMI of at least 30) and 61% of the participants do not reach the minimal amount of physical activity as suggested by the World Health Organization.