Background
The Cooperative Health Research In South Tyrol (CHRIS) study is a population-based study with a longitudinal lookout established in 2011 to investigate the genetic basis of common chronic conditions associated with human ageing, and their interaction with life-style and environmental factors in the general population of South Tyrol. Located in the middle of the Alpine mountainous region, the landscape is characterized by rural and small villages across many valleys. Agriculture and tourism are the major drivers of the gross domestic product of the region, which is among the highest in Europe. In a context of a general worldwide life expectancy increase, ageing is expected to be longest in such high income regions [
1], which will also be characterized by a relevant demographic deficit [
2], as documented in a recent report from the Autonomous Province of Bolzano/South Tyrol [
3].
The CHRIS study is focused on cardiovascular, metabolic, neurological, and psychiatric health. Cardiovascular and metabolic diseases are major components of all non-communicable diseases, whose burden will continue to increase over the next decades [
1]. By 2030, more than 40 % of the adult population could be affected by at least one cardiovascular condition [
4]. Cardiovascular diseases (CVD) will become the leading cause of death in 65 + -year-old subjects and, at the current ageing rate, the cost to treat CVD will triplicate [
5]. Similarly, prevalence of metabolic syndrome [
6], diabetes, and particularly prediabetes [
7], is rising. Among the neurological conditions, the most prevalent disease in European regions of low child and adult mortality is migraine [
8]. However, the increased burden of neurological diseases over the next fifteen years will be mainly due to the rising prevalence of dementias [
9] and neuropathies. The CHRIS study pays particular attention to neurodegenerative movement disorders, since they are generally under-recognized and under-treated, with a relevant impact on life quality and health economy. A South Tyrolean study representative of the general population [
10] reported a prevalence of ~ 28 % in 50–89-year-old subjects, sharply increasing with age. Psychiatric health is also a major concern of European countries, with 14 % of the population suffering of anxiety disorders and nearly 7 % of major depression, with no particular differences by country [
11].
Biomedical research is needed to identify factors that affect the ageing process, which may lead to preventive interventions for healthy ageing with reduction of health care related costs. For this reason, the CHRIS study was established as a collaboration between a research institute (the EURAC Center for Biomedicine) and the South Tyrolean Health System. Such a collaboration guarantees that the study operates by actively interacting with the local population, thus raising awareness towards a more conscious approach to health. The study is expected to foster a dynamic cycle among scientists, clinicians, and the whole population, which is to offer reciprocal feedback to ultimately improve individuals’ health.
Results
In 2015, we froze the first data release which included 4979 participants (male:female ratio = 44 %:56 %) recruited until 15 July 2014. Participants’ characteristics by gender are given in Table
4. On average, subjects were 46.2 years old (SD = 16.4) with no difference between sexes. Ninety-four percent of subjects was born in South Tyrol. A diploma from a vocational school was the most common educational level attained, with females having a higher educational level than males. Eighteen percent males and 5.9 % females reported daily alcohol consumption. Current smokers were 19.7 % of males and 16.7 % of females (
P < 0.01). The majority of the sample was classified as doing high physical activity. Doctor-diagnosed hypertension was reported by 25.1 % males and 21.5 % females (
P < 0.01) and 60.7 % males and 41.3 % females had a BMI > 25 (
P < 0.01). When defining diabetes status based on serum glucose and glycated hemoglobin levels (see Table
4 footnote), 4.6 % males and 4.2 % females were classified as having diabetes, while a pre-diabetic condition was identified in 40.4 % males and 41.7 % females (
P value 0.55). Finally, females had a slightly higher total cholesterol level than males (
P < 0.01). Table
5 shows that the age and sex structure of the recruited sample is very similar to that of the general population of the region. An exception is the 75+-year-old group, which is underrepresented in the CHRIS study due to the logistic difficulty of very old people to travel to the study center. However, in general such similar distribution between sample and reference population suggests that the CHRIS study sample will be valuable also for epidemiological investigations. As an additional characterization of this sample, we observed 606 singleton and 4373 related subjects out of the 4979 participants. The related subjects could be connected through 186 pedigrees characterized by 3014 founders, each one with 1–56 descendants (mean 6). Finally, 4570 samples have been fully genotyped and are available for genome-wide association studies. This dataset will constitute the first nucleus for epidemiological investigations and genetic data analysis.
Table 4
Characteristics of the 4979 participants included in the first CHRIS data release by gender
Age (years)—mean (SD) | 46.5 (16.4) | 45.9 (16.3) | 0.22 |
Self-reported origin—n (%) | | | |
South Tyrol | 2087 (94.4) | 2591 (93.6) | 0.40 |
European | 122 (5.5) | 174 (6.3) |
Non-European | 3 (0.1) | 2 (0.1) |
Education, n = 4976—n (%) | | | |
No formal education or degree | 5 (0.2) | 7 (0.3) | <0.01 |
Primary school | 223 (10.1) | 379 (13.7) |
Lower secondary school | 311 (14.1) | 401 (14.5) |
Vocational school | 1070 (48.4) | 985 (35.6) |
Upper secondary school | 427 (19.3) | 703 (25.4) |
University | 176 (7.9) | 289 (10.5) |
Alcohol consumption frequencyc, n = 4974—n (%) | | | |
Never to seldom | 489 (22.1) | 1389 (50.3) | <0.01 |
≤1 day/week | 684 (30.9) | 897 (32.5) |
≥2 days/week | 630 (28.5) | 313 (11.3) |
Daily | 408 (18.5) | 164 (5.9) |
Smoking habits, n = 4977—n (%) | | | |
Never smoker | 1024 (46.3) | 1550 (56.1) | <0.01 |
Past smoker | 752 (34.0) | 753 (27.2) |
Current smoker | 436 (19.7) | 462 (16.7) |
Physical activity, n = 4683—n (%) | | | |
Low | 502 (24.2) | 626 (24.0) | <0.01 |
Moderate | 507 (24.5) | 853 (32.7) |
High | 1063 (51.3) | 1132 (43.4) |
Blood pressure (BP), n = 4971—mean (SD) | | | |
Systolic BP (mm/Hg) | 124.5 (14.5) | 115.3 (17.1) | <0.01 |
Diastolic BP (mm/Hg) | 79.3 (9.0) | 76.6 (9.5) | <0.01 |
Elevated blood pressured, n = 4953—n (%) | | | |
No | 1647 (74.9) | 2163 (78.5) | <0.01 |
Yes | 551 (25.1) | 592 (21.5) |
Body-Mass-Index kg/m2, n = 4916—n (%) | | | |
<25 | 856 (39.3) | 1607 (58.7) | <0.01 |
25–30 | 972 (44.6) | 701 (25.6) |
≥30 | 350 (16.1) | 430 (15.7) |
Diabetes statuse, n = 4967—n (%) | | | |
No | 1215 (55.0) | 1491 (54.0) | 0.55 |
Prediabetes | 891 (40.4) | 1151 (41.7) |
Diabetes | 102 (4.6) | 117 (4.2) |
Serum total cholesterol (mg/dl)—mean (SD) | 207.8 (41.7) | 213.0 (40.3) | <0.01 |
Table 5
Age and sex distribution of CHRIS participants compared to the general population of Vinschgau/Val Venosta
18–29 | 442 (20.0) | 576 (20.8) | 1018 (20.5) | 2834 (19.9) | 2661 (18.7) | 5495 (19.3) |
30–44 | 583 (26.4) | 739 (26.7) | 1322 (26.6) | 3864 (27.1) | 3659 (25.7) | 7523 (26.4) |
45–59 | 708 (32.0) | 879 (31.8) | 1587 (31.9) | 4001 (28.0) | 3739 (26.3) | 7740 (27.2) |
60–74 | 370 (16.7) | 441 (15.9) | 811 (16.3) | 2340 (16.4) | 2358 (16.6) | 4698 (16.5) |
75+ | 109 (4.9) | 132 (4.8) | 241 (4.8) | 1238 (8.7) | 1803 (12.7) | 3041 (10.7) |
Overall | 2212 (100) | 2767 (100) | 4979 (100) | 14,277 (100) | 14,220 (100) | 28,497 (100) |
Discussion
Biomedical research is witnessing a paradigm shift from sickness to health, where the best way to prevent disease onset is to understand how the whole ageing process works in healthy individuals. For this purpose, longitudinal studies based on the general population are of the greatest relevance. In addition to screening population health through extensive interviews and clinical examinations, the CHRIS study has some distinctive features.
The target population in this restricted area of the Alps is characterized by a rather homogeneous life-style. Previous work from our group demonstrated that the population in this area is stable, with low residential mobility across generations [
12,
57,
58] and low inbreeding [
59]. We previously observed a low to null impact of shared environmental components on biomarker heritability [
13], which suggests homogeneous life-style and environmental conditions. Such a homogeneity would constitute an advantage in mapping causal genetic variants in this geographical area. The collection of non-genetic exposures such as diet, smoking, and physical activity will allow us to assess the extent of variability of life-style in this region and to test candidate gene-environment interactions for biomarker traits of interest and the most prevalent disease outcomes.
The reconstruction of familial information will allow the building of medium sized pedigrees which will facilitate the mapping of inherited variants. With the emphasis currently being on low frequency genetic variants, identified through SNP arrays or next-generation sequencing technology, sampling a large set of the resident population in a small region is advantageous in terms of allowing reliable identification of alleles that may be rare elsewhere but possibly overrepresented in the region. While the level of enrichment of low frequency variants may not be expected to be as large as that of populations undergoing strong genetic isolation [
60], it is likely that the CHRIS population sample could show more enrichment than general population studies. The absence of founder effects may keep the effective population size of the study large [
61] which, paired with environmental homogeneity, should translate into higher power of detecting genetic associations. The study is planned to recruit about 10,000 subjects, corresponding to roughly 35 % of the entire population in the region. Despite the typical issues of Alpine valleys in respect to traveling time and conditions necessary to reach the study center especially from the more remote villages, we believe this rate is reasonably achievable.
Currently, the genetic epidemiology community is facing the issue of data sharing and harmonization. Data harmonization, pooling, and sharing are beneficial to scientific research provided that data security and privacy are guaranteed [
62]. This aspect may be particularly relevant now that rare genetic variants are being considered more frequently, given their functional nature, and data-sharing might become the only way to increase study power. In fact, population-based research is under two opposite ethical pressures: on the one hand it is imperative to protect data privacy and security, on the other hand, an ethically responsible research should aim at maximizing the use of stored data and samples, so as to guarantee the maximal benefit to the community [
16]. For this reason, with the aim of maximizing the scientific harvest of the CHRIS study, collaborations that are instrumental to data harmonization within the Italian Hub of Population Biobanks [
63] and the Biobank Standardisation and Harmonisation for Research Excellence in the European Union (BioSHaRE) initiative [
62,
64] were set up. In particular, the BioSHaRE project framework allows sharing data analysis results without real sharing of the personal-level data, thanks to the new DataSHIELD technology [
65].
A relative advantage of a population-based study of limited sample size conducted in a small region might be the quicker turn-around to establish focused research studies nested into the CHRIS study. The strict collaboration with GPs and hospital doctors from the valley’s central hospital made it easier to build up a trust relationship, so that re-invitation of participants for additional deep phenotyping has been proven to be a realistic target. Along the same line, from a biobanking perspective, such a manageable sample size opens up the possibility of creating a very thick layer of phenotypes from the collected biomaterial. Projects are already ongoing that will establish a broad range of molecular phenotypes covering a wide spectrum of metabolites through LC–MS analysis, effectively extending our previous work [
14,
66]. In this context, small studies are particularly suitable for gene discovery, given that essential metabolites are generally regulated by a very few genes with large effects [
14,
66].
Europe is under an unprecedented ageing phase, which brings the need of a global initiative to prevent morbidities typical of the elderly [
67]. A global approach includes health promotion among the most important policy stakeholders. By enrolling a large portion of the total population and directly involving all GPs and municipality councils of the region, the CHRIS study acts as an important player in raising the attention towards healthier ageing. In addition to the unavoidable sensitization which passes through invitation and participation in our screening program, participants are explicitly recommended to visit their reference GPs for interpretation of the clinical examination results. Informative events are planned in collaboration with local charities promoting knowledge and prevention of the diseases, so that CHRIS researchers are involved in health-promoting events which go beyond the communication about the study. Current efforts are being dedicated to create groups of local promoters of healthy ageing studies, which may help in sensitizing the population towards participation and may also have a role in highlighting special population needs so as to orient research towards applied solutions that are closer to the population needs.
To propose a concept where health can be shaped together by people and scientists, a general initiative was started, where the representatives of the largest associations of the valley together with GPs and local schools are being involved. This group has the purpose to promote health-related initiatives in the valley, increase the level of attention towards health-related topics, and hopefully create a philanthropy culture which can not only support, but foster scientific biomedical research in the region.
In a time when very large biobanks have been established [
68], strict scientist-population turnaround may be an advantage for mid-sized studies in order to achieve phenotype coverage of the target population, precision and homogeneity of the phenotype measurements, and support from the study participants themselves to expand the study into novel areas of biomedicine. On these premises, the CHRIS study will serve as foundation for a larger biomedical research effort aimed at uncovering genetics and molecular markers that can be predictive of individuals’ ageing trajectories as well as serve as potential therapeutic targets in disease progression control [
69].
Authors’ contributions
CP, MG, DM, RM, CS, ADG, LF, YDE, CM, SZ, SVS, AR, AAH, and PPP drafted the manuscript; MFF, TS, and PPP planned and designed the study; CP, MG, CE, SZ, YDE, AR, SVS, DM, and ADG designed and developed specific modules within the study; BL, MFF, and CM planned and performed the feasibility study; CP, MG, DM, RM, CS, ADG, LF, YDE, CF, CM, CE, LSK, SZ, MFF, SVS, AR, AAH, HW, and PPP critically reviewed the manuscript.
Acknowledgements
The CHRIS study is a collaborative effort between the Center for Biomedicine of the European Academy of Bolzano/Bozen (EURAC) and the Healthcare System of the Autonomous Province of Bolzano (Südtiroler Sanitätsbetrieb/Azienda Sanitaria dell’Alto Adige). The CHRIS Study is affiliated to the “German National Cohort” (Germany) and is indebted with the investigators of this study for their support in the study protocol definition.
We thank all inhabitants of the middle and upper Vinschgau/Val Venosta; Dr. Florian Wöhs and Dr. Stefan Haumer, for their support in the ECG analysis, and the personnel of the hospital of Schlanders/Silandro, directed by Dr. Anton Theiner; Dr. Stefan Platzgummer, Dr. Monika Alber, Flora Gnech, and all the personnel of laboratory medicine department of the hospital of Meran/Merano for their collaboration of the biochemical laboratory and fundamental support with biosample analysis; Dr. Oswald Mayr, director of the South Tyrolean Healthcare Service, for the continuous support of the study; the general practitioners: Stefan Waldner, Wunibald Wallnöfer, Raffaela Stocker, Josef Stocker, Bettina Skocir, Helmut Rauner, Georg V. Hofer, Christian Hofer, Erich Dona’, Hansjörg Gluderer, Robert Kaserer, Ugo Marcadent, Anton Pizzecco, Josef Plangger, Monica Scherer, Oswald Tappeiner, and Günther Bauer; the study assistants and nurses at the study center: Roselinde Gunsch, Karin Bystrianska, Brunhilde M. Grasser, Benedikta Linter, Lea E. Moriggl, Liane Parth, Susanne Saewert, and Renate Telser (former members: Tamara Oberhofer, Marilena Koch); the laboratory technicians: Ilaria Bozzolan and Giulia Caprioli (former members: Simona Amistadi, Lisa Longaretti, Marlene Obkircher, Stefanie Wieser); the IT data manager: Johannes Martin and Daniele Di Domizio (former member: Daniel S. Schmitt); Andreas Wiedmer for his technical support; Larissa De Clauser for IT support and quality controls; Arne Pfeufer and Simon Rauch for helping with protocol development; the administrative collaborators: Vera Amon, Vanessa Leitner, and Marika Fregnan (former member: Andrea Vieider); Massimiliano Pellegrini for the legal support; the members of the CHRIS scientific advisory board: Ettore Beghi, Maria B. Donati, Wolfgang Lieb, Christine Meisinger, and Barbara Parodi; the members of the CHRIS evaluation committee: Dr. Arno Gasperi, Dr. Hermann Brugger, and Edmund Gurschler.
We thank: Drs. Georg Schmidt (Technical University of Munich) and Daniela Berg (University of Tübingen) for helpful discussion; Drs. Markus Nöthen, Per Hoffmann, and Stefan Herms (University of Bonn) for the genotyping; Drs. Marcella Rietschel, Fabian Streit, Jana Strohmaier, and Stephanie Witt (University of Mannheim) and Dr. Ettore Favaretto (Hospital of Brixen/Bressanone) for help in developing the psychiatric assessment; Drs. Roberto de Marco (University of Verona) and Joachim Heinrich (Helmholtz Zentrum München) for permission to use the ECRHS smoking questionnaire; Dr. Vanessa Garcia-Larsen for support in implementing the GA2LEN FFQ; Dr. Ruth Ruscheweyh (University of Munich) for supervising the implementation of the pain sensitivity analysis.