Burden of non-communicable diseases in Cyprus, 1990–2017: findings from the Global Burden of Disease 2017 study

Background Non-communicable diseases (NCDs) accounted for over 90% of all deaths in the Cypriot population, in 2018. However, a detailed and comprehensive overview of the impact of NCDs on population health of Cyprus over the period of 1990 to 2017, expressed in disability-adjusted life years (DALYs), is currently not available. Knowledge about the drivers of changes in NCD DALYs over time is paramount to identify priorities for the prevention of NCDs in Cyprus and guide evidence-based decision making. The objectives of this paper were to: 1) assess the burden of NCDs in terms of years of life lost (YLLs), years lived with disability (YLDs), and DALYs in Cyprus in 2017, and 2) identify changes in the burden of NCDs in Cyprus over the 28-year period and assess the main drivers of these changes. Methods We performed a secondary database descriptive study using the Global Burden of Disease (GBD) 2017 results on NCDs for Cyprus from 1990 to 2017. We calculated the percentage change of age-standardized DALY rates between 1990 and 2017 and decomposed these time trends to assess the causes of death and disability that were the main drivers of change. Results In Cyprus in 2017, 83% (15,129 DALYs per 100,000; 12,809 to 17,707 95%UI) of total DALYs were due to NCDs. The major contributors to NCD DALYs were cardiovascular diseases (16.5%), neoplasms (16.3%), and musculoskeletal disorders (15.6%). Between 1990 and 2017, age-standardized NCD DALY rates decreased by 23%. For both males and females, the largest decreases in DALY rates were observed in ischemic heart disease and stroke. For Cypriot males, the largest increases in DALY rates were observed for pancreatic cancer, drug use disorders, and acne vulgaris, whereas for Cypriot females these were for acne vulgaris, psoriasis and eating disorders. Conclusion Despite a decrease in the burden of NCDs over the period from 1990 to 2017, NCDs are still a major public health challenge. Implementation of interventions and early detection screening programmes of modifiable NCD risk factors are needed to reduce occurrence and exacerbation of leading causes of NCDs in the Cypriot population. Supplementary Information The online version contains supplementary material available at 10.1186/s13690-021-00655-8.


Background
Non-communicable diseases (NCDs) are a major cause of death and disability worldwide [1]. The four major groups of NCDs, namely cardiovascular diseases (CVDs), neoplasms, chronic respiratory diseases, and diabetes, cause over 70% of global mortality each year [1]. In Cyprus, a high-income country and a member state of the European Union (EU), NCDs have posed an even greater burden on population health. A 2018 World Health Organization (WHO) report revealed that NCDs accounted for over 90% of all deaths in the Cypriot population [1]. Markedly, neoplasms accounted for 19% of all deaths in Cyprus [2]. Moreover, the prevalence of diabetes in the Cypriot population was estimated to be 10% which is slightly higher compared to the prevalence of diabetes in other EU countries [3]. Additionally, Cyprus has the highest relative disease burden from diabetes (4.6%), although publishes proportionally less (1.2%) with almost half of research output dedicated to Type 2 diabetes, compared to the rest of the European countries (including the UK) [2]. Furthermore, smoking, physical inactivity and childhood obesity are some of the main public health risks identified in previous studies [2,4,5] as contributing to the growing NCD epidemic. Cyprus has a relatively centralized and co-ordinated vaccination system at the national level, with programmes targeting a covering at nearly 100% [6]. Therefore, the contributing burden of disease on mortality and morbidity are heavily based on NCDs. WHO member states have endorsed a set of policy options and cost-effective NCD interventions that can be used to tackle the burden of NCDs [7]. Furthermore, population ageing is causing an increase in the burden of specific health conditions; WHO's Global Strategy and Action Plan on Ageing and Health 2016-2020 [8] urges countries to establish intervention policies on healthy population ageing. Understanding at which ages the NCD burden starts to accumulate may shed light on how to introduce better policies and hence, how to reduce the projected NCD burden. The economic growth and prosperity in Cyprus (ranked 33th out of 167 countries) with health and living conditions being at the top of the indicators list, may be responsible for the reduction of poverty [6], but may have contributed to the increase of obesity as a sideeffect of such a wealthy lifestyle. To exacerbate, as Cyprus has life expectancy at 82.2 years of age (amongst the highest life expectancy at birth between European countries) [9], the ageing population is more likely to have more than one chronic disease, with further imposed health burdens for the individual and healthcare associated costs to the society.
However, resources are limited and it is therefore important for policy makers to have up to date quantifications of the impact of NCDs on the population health of Cyprus and the relative attributes of modifiable risk factors to guide priority setting.
The impact of NCDs on population health can be quantified using mortality or incidence and prevalence. However, NCDs are characterized by heterogeneity in health outcomes, with great variety in severity, duration and mortality rates [10]. Moreover, with increasing life span, information on disability has become more important. Summary measures of population health (SMPH) combine mortality, morbidity and disability into one single index [11,12]. This allows for comparison between distinct health outcomes, and subsequently, comparison of the population health impact of a range of diseases and risk factors. SMPHs are therefore vital tools for priority setting purposes [11,12]. A widely used SMPH is the Disability-Adjusted Life Year (DALY) [10]. The DALY-concept integrates premature mortality in years of life lost (YLLs) and morbidity in years lived with disability (YLDs) [11,12]. The DALY has been used in the landmark Global Burden of Disease (GBD) studies which aim to assess up to date country-specific incidence, prevalence, mortality, YLLs, YLDs, and DALYs for over 300 diseases and injuries in 195 countries and territories using a systematic analysis [13][14][15]. The approach that is used by the GBD researchers ensures that the calculated incidence, prevalence, mortality, YLLs, YLDs, and DALYs are comparable and internally consistent across years and regions.
Annually, updated methods and results of the GBD study are published. However, a detailed and comprehensive overview of the burden of disease of NCDs in Cyprus is currently not available. Investigation of time trends and decomposition of these time trends can pinpoint the diseases that contributed most to the change in burden of disease of NCD. Knowledge about the main drivers of changes in NCD YLDs, YLLs, and DALYs over time is imperative for health professionals and policy makers to identify priorities for the prevention of NCDs in Cyprus and guide evidence-based decision making. Until recently, the health authorities in Cyprus lacked knowledge about the population's state of health and health policies have not been targeted. In fact, it was shown that the biomedical research funded on the island does not correspond to the DALYs of the Cypriot population [2]. Therefore, assessing the burden of NCDs in Cyprus is an important topic for public health policy planning for primary health interventions and prioritizing NCD prevention policies.
Here, we have sought to provide a comprehensive overview of the age-standardized YLLs, YLDs and DALYs of NCDs in Cyprus in 2017, to investigate rates of change over the period from 1990 to 2017, and to identify the NCDs that were main drivers of these changes. Comparison of NCD-related DALYs between age-specific and age-standardized rates was also made.

Overview
We performed a secondary database descriptive study using the GBD 2017 results. The GBD 2017 study analyzed the impact of 359 diseases and injuries across 23 age groups and both sexes, and 195 countries and territories between 1990 and 2017 [15]. Detailed descriptions on the GBD study methodology, data, and analysis have been previously described [15,16]. In this study we restricted our analysis to YLLs, YLDs, and DALYs due to NCDs in the Cypriot population from 1990 to 2017. Briefly, YLLs are calculated by multiplying the number of deaths by the global standard life expectancy at that age. YLDs are calculated as prevalence of a health state multiplied by the corresponding disability weight of this health state. YLD estimates are corrected for comorbidity using methods described elsewhere in the GBD study [15]. DALYs are calculated by adding the YLLs and YLDs, thereby incorporating both mortality and morbidity [13][14][15][16]..
The diseases studied by GBD are arranged in standard hierarchical categories of four levels. Level 1 causes consist of three categories, namely: communicable, maternal, neonatal, and nutritional diseases (Group I); NCDs (Group II); and injuries (Group III). Each Level can be broken down into a more detailed classification. For example, Group II can be broken down into 12 different diseases at Level 2. The NCD categories (at Level 2 of the cause hierarchy) featured in the GBD 2017 study were neoplasms, cardiovascular diseases, chronic respiratory diseases, digestive diseases, neurological disorders, mental disorders, substance use disorders (SUDs), diabetes and chronic kidney disease (CKD), skin and subcutaneous diseases, sense organ diseases, musculoskeletal disorders, and other NCDs. If the interest is for DALYs on CVDs, for example, then these causes can be further broken down into 11 sub-diseases (Level 3) and more details (Level 4). The GBD 2017 disease categories by level can be found elsewhere [17]. For the present analysis, we report Level 2 and Level 3 NCD (sub-)causes (see Additional file 1).
Cyprus has been divided into two parts; the northern part which is under Turkish occupation and the southern or government-controlled part which consists of five districts, namely Nicosia, Ammochostos, Larnaca, Limassol, and Paphos. The total population in the government-controlled area is estimated at 875,900 in 2018 [18]. All districts of Cyprus will be referred to as a whole in the rest of the manuscript.

Source of data and presentation
In our study, we analyzed and reported levels and trends of age-standardized YLL, YLD, and DALY rates. An agestandardized rate is a weighted average of the age-specific rates per 100,000 of population, where the weights are the proportions of the standard population in the corresponding age groups. Cause-specific mortality was informed primarily from vital registration data. Epidemiological data from scientific reports and health surveys were used to generate NCD-specific prevalence and incidence estimates. YLL and YLD estimates due to NCDs were adjusted for incompleteness and misclassification using standardized approaches [15,16]. YLLs, YLDs, and DALYs were provided by the visualisation "GBD Results" tool (Institute for Health Metrics and Evaluation (IHME), 2017; available online at: http:// ghdx.healthdata.org/gbd-results-tool).
Age-standardized DALY rates, and its components YLL and YLD, related to all NCDs in Cyprus from 1990 to 2017, were analyzed for both genders at Levels 2 and 3. We, first, calculated the proportion of the top five Level 2 and Level 3 NCDs that contributed most to the burden of NCD DALYs, YLLs, and YLDs in Cyprus in 2017 using the cross-multiplication method, known as 'rule of three' [19]. For changes over time, we presented the percentage of change for each age-standardized YLL, YLD, and DALY rate in 1990 and 2017. A positive percentage change indicates an increase, whereas a negative change a decrease from 1990 to 2017. Additionally, we decomposed differences in the DALY-related NCDs over the period from 1990 to 2017 to assess the main drivers of change in DALY, YLL and YLD rates of NCDs. Finally, we have examined the age distribution of NCD DALYs and we reported the NCD DALY rates for the elderly (70+ years) category over the 28-year study period.

Uncertainty
Uncertainty distribution for each NCD outcome variable (YLL, YLD, and DALY) was captured and propagated by 1000 draws from the posterior distributions. The results for each variable of interest were derived from the mean of 1000 draws and the 95% uncertainty intervals (UIs) were derived from the 2.5th and 97.5th percentiles of the corresponding draws of the sampled YLL, YLD, and DALY variables [20,21].

Burden of disease of NCDs, in Cyprus in 2017
In Cyprus in 2017, the total burden of disease was 18, 287 DALYs (15,607 to 21,322 95%UI) per 100,000 of which 83% (15,129 DALYs per 100,000; 12,809 to 17,707 95%UI) were due to NCDs.
Overall, the NCD DALY rates per 100,000 were similar in males (16,627 DALYs per 100,000; 14,357 to 19, 159 95%UI) and females (13,923 DALYs per 100,000; 11, 453 to 16,802 95%UI). However, in the older age categories from age 50 and older the higher NCD DALY rates among males compared to females stand out. In addition, Fig. 1 Table 1 shows the rankings for the top five Level 2 and Level 3 NCDs that contributed most to overall NCD YLLs, YLDs, and DALYs in Cyprus in 2017.
Changes in NCD DALY rates in Cyprus, 1990-2017 Table 2 shows the age-standardized YLL, YLD, DALY rates and percentage change for Level 2 NCD-group, in Cyprus between 1990 and 2017.
Over the period from 1990 to 2017 NCD DALY rates in Cyprus decreased from 19,608 DALYs per 100,000

Changes in NCD DALY rates in the elderly (70+ years) in Cyprus, 1990-2017
Over the period from 1990 to 2017, the NCD DALY rates in the elderly (70+ years) were higher compared to the age-standardized NCD-related DALYs per 100,000 of other age groups. Major contributors to the NCD DALYs in elderly in 2017 were CVDs, neoplasms, and diabetes and CKD. CVDs accounted for 27.5% (20,890 DALYs per 100,000; 19,253 to 22,783 95%UI), neoplasms for 18% (13,950 DALYs per 100,000; 12,687 to 15,184 95%UI), and diabetes and CKD for 10% (7761 DALYs per 100,000; 6808 to 8847 95%UI) of the total NCD burden. Leading Level 3 causes of NCDs in the elderly in A similar pattern was also seen in DALYs due to diabetes; from 11,285 DALYs per 100,000 (9976 to 12,614 95%UI) in 1990 to 7761 DALYs per 100,000 (6808 to 8847 95%UI) in 2017, representing a decrease of 31%. More gradual decreases in NCD DALY rates among the elderly population were observed for neoplasms, digestive diseases, mental and neurological disorders, chronic respiratory diseases, and other noncommunicable diseases. Figure 3 shows the NCD DALY rates in the elderly population over the period from 1990 to 2017.

Summary of findings
The findings of this study showed that in Cyprus in 2017 83% of the total burden of disease was attributable to NCDs, and that CVDs, neoplasms and musculoskeletal disorders were the top contributors to the burden of NCDs. Between 1990 and 2017, age-standardized NCD DALY rates decreased by 23%. For both males and females, the largest decreases in DALY rates were observed in IHD and stroke. However, over this 28-year period, CVDs, neoplasms, and musculoskeletal disorders were consistently major contributors to NCD DALYs for both males and females. In particular, neoplasms and CVDs were mostly driven by YLLs, whereas musculoskeletal disorders were driven by YLDs. From 1990 to 2017, NCD DALY rates were highest among the elderly (70+ years).
Although the NCD burden of IHD and stroke, as quantified by age-standardized DALY rates, declined for both Cypriot males and females over the period from 1990 to 2017, they remained the main CVD DALY contributors. A possible explanation for this may be that around 30% of Cypriots have untreated hypertension [22]. Since hypertension is a major risk factor for IHD and stroke, the high prevalence of hypertension may therefore be a main contributor to the burden of IHD and stroke. Also, smoking is a well-established risk factor for IHD and stroke. Notably, the prevalence of tobacco use over time is consistently high in Cyprus compared with other European countries [23] and high IHD and stroke DALY rates may reflect the legacy of high cigarette smoking rates in Cyprus. To achieve further decreases in the prevalence and mortality of IHD and stroke in Cyprus it is therefore important to target these risk factors. For instance, by setting up early detection screening programmes for hypertension, and scaling up of smoking cessation interventions to prevent the disease burden possibly attributable to the tobacco use among the Cypriot population.
The proportion of DALY for diabetes in Cyprus was higher than in the WHO European Region [2]. Our findings showed that YLDs caused by diabetes increased for Cypriot males as well as females, during the period of 1990-2017. Environmental, lifestyle and genetic factors might have a significant effect on the pathogenesis of diabetes in the Cypriot population [24,25]. Thus, the large proportion of diabetes-related DALY in the total NCD burden, as found in this study, calls for efforts to investigate leading risk factors for diabetes and kidney disease in Cyprus as well as improvements in diabetes management.
Cypriot males had substantially more YLLs and DALYs due to pancreatic cancer than females. Alcohol consumption and tobacco use have been identified as major risk factors for pancreatic cancer [26,27]. In Cyprus, the smoking prevalence is high; similarly, the prevalence of heavy episodic drinking is estimated to be 28%, which is close to that of WHO EU countries (30%) [28]. However, a high proportion of alcohol-related and smoking-related disease burden increase the risk of other health conditions, such as CVDs and neoplasms.
Smoking cessation and alcohol abuse interventions have been shown to be effective with a potential effect on public health [29,30]. Thus, alcohol and tobacco control policies should also be considered in the Cypriot primary health care. Also, during the period from 1990 to 2017, the burden of SUDs increased for Cypriot males. This burden is mainly driven by alcohol and drug use disorders. Previous studies have yielded evidence of an association between alcohol consumption and/or drug use and unemployment status among males [31,32]. Between 1990 and 2017, the unemployment rate of the labor force in Cypriot males was increased from 1.4 to 10.4% whereas in Cypriot females from 2.5 to 9.8% [33]. According to the 2003 and 2019 EU Health Surveys, the prevalence of smoking among Cypriot males decreased slightly from 38 to 32%, respectively. On the other hand, the prevalence of smoking for Cypriot females increased from 10% (in 2003) to 13% (in 2019) [34,35]. In addition, the prevalence of heavy episodic drinking among the Cypriot population was decreased by 8% (2010-2016) [28]. This highlights the importance of both strengthening the social welfare policies as well as incorporating such policies in population mental health promotion strategies.
Age-standardized YLD and DALY rates due to eating disorders, such as anorexia nervosa and bulimia nervosa, have been identified as being higher in Cypriot females than males. The burden of eating disorders has traditionally been linked to body-image dissatisfaction and the role of social media [36,37]. More research is needed in order to explore specific determinants for eating disorders in the Cypriot community. Nonetheless, the development and use of validated screening tests in primary healthcare setting may help to determine future health strategies regarding the burden of eating disorders in Cyprus.
Between 1990 and 2017, the leading cause of DALYs in elderly (70+ years) shifted to CVDs. A possible explanation for this is that most of the metabolic risk factors namely high fasting plasma glucose and/or high blood pressure are highly prevalent in the aged [38]. Over the 28-year study period, DALYs due to psoriasis in the elderly increased by 17%. The etiology of psoriasis involves interaction between genetic factors and exposure to smoking, alcohol drinking, and unhealthy dietary habits [39]. The association between CVDs and psoriasis and increased prevalence of cardiovascular risk factors have been described elsewhere [40]. The increasing impact of psoriasis among the elderly in Cyprus may be explained primarily by the shift towards unhealthy dietary habits and the high prevalence of NCDs, over the last decades. The interaction of multiple health conditions and risk factors prove a challenge for the prevention of NCDs in elderly. However, CVD risk assessment in elderly emerges an essential priority for health policy authorities.

Strengths and limitations of the study
This study has several strengths and limitations. The present study has introduced the use of GBD 1990-2017 results to provide a comprehensive, up-to-date and indepth overview of the burden of NCDs in terms of YLL, YLD, and DALY in Cyprus. A major strength of the GBD estimates is the internal consistency and comparability of the age-standardized YLL, YLD, and DALY estimates, which allow comparison across various countries and regions at multiple time points. Therefore, our findings are indispensable in helping Cypriot policymakers to develop evidence-based prevention and intervention strategies for NCDs.
This study shares the limitations of the GBD 2017 study, which have been discussed in detail elsewhere [15,41]. First and foremost, the GBD methodology produces sub-regional estimates for a number of countries; however, sub-regional estimates are not available for Cyprus.
For Cyprus this means that the estimates presented here are based on the northern and the southern part of Cyprus combined. However, since the northern part is under Turkish occupation and the southern part is controlled by the government of Cyprus, there may be differences in health policy and prevention measures in the northern and southern part, which may impact NCD DALY rates and trends over time. Due to the unavailability of estimates for Cyprus, sub-regions variability across these sub-regions cannot be studied. Similarly, the GBD does not produce estimates for sub-groups of the population according to, for example, socioeconomic status or ethnic background. Second, we did not analyze burden of NCDs by age groups, other than the 70+ age group, in detail, which would be necessary for the implementation of age-specific intervention strategies and/or activities in primary healthcare in Cyprus. Third, the NCD prevalence data in Cyprus is limited; the DALY estimates, in the GBD study, were informed by data from 32 data sources that consisted of health survey data and scientific literature reports. Both the low quality as well as possible low quality of these data sources may have introduced uncertainty and possible led to large UIs. The causes-of-death data in Cyprus, on the other hand, are predominantly provided by vital registration system. According to the GBD standard procedure, cause-of-death data are coded based on the International Classification of Disease (ICD) rules and the miscoded and non-specific coded deaths are re-assigned to specific cause-of-death categories. Therefore, there may be differences in the number of deaths by cause-of-death reported in the GBD 2017 study and those reported by the Cyprus statistical services.

Implications for health policy in Cyprus
Our findings have important implications for evidencebased decision-making on the NCD intervention strategies in Cyprus. The majority of the NCDs share modifiable risk factors, namely tobacco use, hypertension, unhealthy diets, and alcohol abuse. Policymakers in Cyprus should consider targeting these NCD risk factors in targeted health prevention policies. Cyprus has a similar NCD risk profile compared with other Mediterranean countries (Greece, Italy, France, Spain, etc.) and the effect of preventive policies mainly for tobacco control has been evaluated [42,43]. From this perspective, planning and developing patient-centered interventions of the NCD risk factors and/or early detection and disease screening can reduce the incidence of NCDs and exacerbation of prevalent NCDs. More importantly, the Cypriot health authorities should formulate and enact on prevention and health promotion strategies for NCDs in order to reduce population exposure in NCD risk factors. Furthermore, there is need to strengthen the epidemiological base for NCD prevalence in Cyprus.

Conclusions
Despite a decrease in the burden of NCDs over the period from 1990 to 2017, NCDs are still a major public health challenge with CVDs, neoplasms, and musculoskeletal disorders to be major contributors to the burden of NCDs for both males and females. Implementation of early detection screening programmes of modifiable NCD risk factors and population-level health promotion programmes are needed to reduce the incidence and exacerbation of leading causes of NCDs in the Cypriot population.