Mortality data for the 20 English- or Dutch-speaking Caribbean countries and territories were obtained from the Caribbean Public Health Agency (CARPHA). Data collection processes in these countries differ from country to country. However, in all of these English- or Dutch-speaking Caribbean countries, medical physicians complete the medical cause of death certificate and trained coders assign codes from the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10) to the conditions listed and select the underlying cause of death using World Health Organization (WHO) rules and guidelines. Data submitted to CARPHA (and its predecessor, the Caribbean Epidemiology Centre (CAREC)) are reviewed and validated before data are accepted.

Data for the 2 United States territories (Puerto Rico and the U.S. Virgin Islands) were obtained from the U.S. Centers for Disease Control and Prevention’s National Center for Health Statistics for the same time frame where both mortality and population data were available.

For each of the 22 countries/territories, we calculated age-standardized mortality rates due to cancer (ICD-10: C00-C97), heart disease (ICD-10: I00-I09, I11, I13, I20-I51), cerebrovascular disease (ICD-10: I60-I69), and diabetes (ICD-10: E10-E14). Mortality and population data from the most recent (at the time of analysis) available 10 years (ranging from 1999 to 2014) were used; data were not available for the same 10 years for each of those countries and territories (16).

We also examined whether there were significant 10-year changes in trends of mortality rates for each of the NCDs and each of the countries/territories. Calculations were completed using SEER*Stat software (17) and the WHO 2000-2025 World Standard Population (18). The annual percent change (APC) for trends was calculated using Joinpoint software (19); joinpoint regression analysis was used to assess changes in linear slope for mortality trends over time. Overall trends in mortality were initially assessed with no joinpoints and tested for significant changes in the model with the addition of one point where there is a significant change in the slope of the line (shown as Trend 1 and Trend 2 later in Table 2). The model also computes an estimated APC for each trend by fitting a regression line to the natural logarithm of the rates.

To calculate regional age-standardized mortality rates, we restricted the time period under review to 2005-2010, for which data were available from 18 of the 20 English- or Dutch-speaking Caribbean countries/territories (all of them except the British Virgin Islands and the Turks and Caicos Islands).

Population data were not available from some of the countries/territories for the entire study period. In in these instances, the most recent year of available data (census or mid-year population estimates) was used to populate subsequent years with missing population data. Age-standardized mortality rates are not presented where there were fewer than six deaths annually, an exclusion that has previously been applied with cancer mortality in the Caribbean (12).

During the period of 1999-2014, four NCDs accounted for 39%-67% of all deaths in the two United States territories and in the other Caribbean countries/territories studied. Those four diseases were also responsible for more than half of the deaths in 17 of the countries/territories. For the combined time period (that is, 10 years for each of the countries/territories), cancer (followed by heart disease) was the leading cause of death for 9 of the countries: Anguilla, Aruba, Barbados, Bermuda, British Virgin Islands, Cayman Islands, Dominica, Jamaica, and Saint Lucia (Table 1).

In the majority of the 20 English- or Dutch-speaking Caribbean countries/territories, heart disease accounted for a higher percentage of deaths (13%-25%) than did cancer (8%-25%), diabetes (4%-21%), and cerebrovascular disease (1%-13%). The U.S Virgin Islands had the highest percentage of deaths due to heart disease (27%) of the 22 Caribbean countries/territories. Among the 20 English- or Dutch-speaking Caribbean countries, the highest percentage of deaths due to heart disease was seen in Montserrat, followed by Bermuda and Trinidad and Tobago. The highest percentages of deaths due to cerebrovascular disease were observed in Saint Kitts and Nevis, Suriname, and Jamaica. The highest percentages of deaths due to diabetes were found in Montserrat, Trinidad and Tobago, and Jamaica. Bermuda, the Cayman Islands, and Aruba had the highest percentages of deaths due to cancer (Table 1).

Figure 1 presents annual age-standardized mortality rates (ASMRs) for each of the countries/territories and each of the four NCDs for each of the years with available data. ASMRs for cancer ranged from 64.7 per 100 000 for Guyana in 2006 to 224.0 per 100 000 for Saint Kitts and Nevis in 2006. ASMRs for heart disease ranged from 48.8 per 100 000 for Turks and Caicos in 2009 to 274.8 per 100 000 for Saint Kitts and Nevis in 2004. ASMRs for cerebrovascular disease ranged from 15.2 per 100 000 for Anguilla in 2009 to 175.5 per 100 000 for Saint Kitts and Nevis in 2004. Finally, ASMRs for diabetes ranged from 10.4 per 100 000 for Montserrat in 2006 to 150.2 per 100 000 for Montserrat in 2011.

Of the group of 20 English- or Dutch-speaking Caribbean countries/territories, all but two of them (British Virgin Islands and Turk and Caicos Islands) provided mortality data for the years 2005 to 2010, for which regional mortality rates were calculated (Figure 2). Overall, ASMRs for all four NCDs remained consistent in this five-year period. However, the gap in mortality between cancer and heart disease closed as the years progressed. When examining regional mortality rates by sex, ASMRs due to cancer and heart disease ranged, respectively, from 131.6 to 141.1 and from 139.5 to 147.4 per 100 000 for males, and from 90.7 to 100.0 and from 97.7 to 108.6 per 100 000 for females. ASMRs due to cerebrovascular disease and diabetes ranged, respectively, from 73.9 to 79.8 and from 68.0 to 77.7 per 100 000 for males, and from 66.7 to 70.5 and from 74.0 to 80.0 per 100 000 for females. Thus, there were no substantial changes in mortality trends in the region from 2005 to 2010 for any of the NCDs (Figure 2).

Investigation into 10-year NCD mortality trends by country/territory revealed some significant trends, as noted in Table 2. (For the majority of the countries/territories, trends remained unchanged over the 10-year study period, thus no values are presented for Trend 2 in those cases in Table 2.). For example, cancer mortality significantly increased in the Bahamas (2004-2008), Jamaica (2002-2011), Saint Vincent and the Grenadines (2005-2014), Suriname (2005-2014), and Trinidad and Tobago (2001-2010). Significant decreases were observed in Belize (2005-2014) and Puerto Rico (2005-2014). Mortality due to heart disease significantly increased in the Bahamas (2004-2009) and Dominica (2005-2014), but significantly decreased in Antigua and Barbuda (2005-2014), Bermuda (2005-2011), Saint Kitts and Nevis (2004-2007), Trinidad and Tobago (2001-2010), Puerto Rico (2005-2014), and the U.S. Virgin Islands (2003-2012). Mortality due to cerebrovascular disease increased significantly in Anguilla (2005-2014), the Bahamas (2004-2008), and Saint Vincent and the Grenadines (2005-2014), but significantly decreased in Bermuda (2005-2014), Suriname (2005-2012), Trinidad and Tobago (2001-2010), Puerto Rico (2005-2014), and the U.S. Virgin Islands (2003-2012). Mortality due to diabetes significantly increased in Grenada (2005-2014) and Jamaica (2002-2011) but decreased in Aruba (2005-2014) and Barbados (2004-2013).

Cancer, cardiovascular disease, and diabetes accounted for more than half the mortality in the English- or Dutch-speaking Caribbean countries/territories in the study. The reasons for the high mortality due to NCDs in those countries/territories remains a critical public health issue that warrants further investigation.

NCD-associated risk factors such as tobacco smoking, harmful use of alcohol, poor diet, and physical inactivity substantially affect NCD mortality (20). Physical inactivity has been associated with an increased risk of certain NCDs, including heart disease, diabetes, and cancer. Available data shows that 31% of the world’s population are not meeting the minimum recommendations for physical activity (21). In 2010, the global prevalence of inactivity in adults aged 18 years and older was 23%. Of the WHO regions, the Region of the Americas (which includes the Caribbean region) had one of the highest prevalence levels of insufficient physical inactivity, at 32% (22).

Harmful use of alcohol has been estimated to be responsible for 3.3 million deaths globally every year, which represent 5.9% of all deaths (23). The highest alcohol consumption levels are found in the developed world, in particular in the WHO European Region and the Region of the Americas. Total alcohol consumption per capita among those 15 years of age and older in the Caribbean countries was more than 7.5 liters of pure alcohol in 2010 (23). Globally, obesity rates have more than doubled since 1980. WHO estimated that in 2014 more than 600 million adults worldwide (~13%) over the age of 18 were obese (22).

Among world regions, the Region of the Americas has one of the highest proportions of adults who are obese, at more than 30%. Based on 2014 data, the Bahamas (36.2%) surpassed the United States (33.7%) (22). Jamaica’s obesity rates are above 30% in those 15 years and older (24).

Latin American and Caribbean nations have a high burden of cardiovascular disease, with 8.7% of total cardiovascular disease deaths being due to high sodium intake (25). A 30% relative reduction in mean population intake of salt/sodium is one of nine global targets for NCDs for 2025 that have been set by WHO (26). PAHO and civil society in Latin America and the Caribbean have taken action to reduce sodium intake, such as with social marketing strategies to reduce salt consumption (27-29).

The findings of our study underscore the importance of implementing evidence-based policies and programs that could mitigate chronic disease risk factors in the Caribbean region (30). Primary health care plays an important role for the prevention of obesity among children and adolescents, such as by promoting breast-feeding, healthy dietary habits, and physical activity. A long period of breast-feeding may reduce the prevalence of obesity among children and mothers (31).

There were shifting mortality trends and great variability observed for specific NCDs within the Caribbean region over the time period that we reviewed. Limited data on NCDs in the region have prevented an accurate assessment of the NCD burden and have created challenges in the development of effective interventions. Many of the Caribbean countries and territories lack standardized surveillance systems for NCDs, with a few exceptions. Barbados performs surveillance for stroke, heart disease, and cancer through the Barbados National Registry (BNR) (32). The cost of this population-based NCD surveillance system is around just US$ 1 per person per year, thus making the case that other small island nations with limited resources should consider similar programs for combined surveillance of NCDs (33). In addition, a few Caribbean countries have implemented high-quality population-based registries for the surveillance of cancers (34). The International Agency for Research on Cancer Caribbean Cancer Registry Hub, implemented under the Global Initiative for Cancer Registration, is working to improve the quality of cancer data obtained from Caribbean cancer registries. However, the cost associated with tracking incident cases from these NCDs is relatively high and presents a challenge for the region (35). Nevertheless, a better understanding of the burden of NCDs will aid in the appropriate allocation of resources needed to address these public health concerns. The capacity to address NCDs is a serious issue and should be assessed carefully. One thing that might be useful for that is a new tool that the U.S. Centers for Disease Control and Prevention has developed to help countries identify key opportunities and gaps in NCD capacity; the tool has been piloted in several countries (36).

In addition to surveillance of NCDs and related risk factors, health sector activities such as screening, diagnosis, and management are also important. Health care systems in many developing countries are not well prepared to deal with a high burden of NCDs, and there is an increased recognition of the need to strengthen primary care services in the Caribbean (37). One strategy that can be used within a primary care setting in the prevention of NCDs is to assess patients for common risk factors and to provide counseling on behavior modification (37). However, a study of primary care performance in Latin America and Caribbean countries identified substantial gaps in primary care, including the lack of health advice offered by physicians to their patients (38). For example, studies done in Caribbean countries confirm that hypertensive and diabetic patients seen in primary care services do not achieve optimal blood pressure and glucose control. These studies indicate that adherence to clinical guidelines for management of hypertension and diabetes may be low among primary care physicians in the Caribbean (39-44). Hypertension is a major risk factor for heart disease and cerebrovascular disease and is also strongly related to diabetes. The Standardized Hypertension Treatment and Prevention Project, which was supported by the U.S. Centers for Disease Control and Prevention and piloted in Barbados, is one example of how it is possible to address the burden of hypertension, by using standardized treatment protocols (45).

In the Caribbean region, NCDs have long been recognized as an area of major concern and one that requires a multisector, regional response (20). One noteworthy move toward that occurred in 2007, with the Caribbean Community’s convening of the world’s first-ever summit of heads of government on NCD prevention and control and the issuance of the landmark Declaration of Port-of-Spain: Uniting to Stop the Epidemic of Chronic NCDs. The 15-point proclamation targets reducing the main NCD risk factors, as well as improving the health service response to NCDs in the region (20). Our analysis in this article is important as a means of monitoring the Caribbean region’s progress towards the Port-of-Spain Declaration targets. Our assessment could also allow individual countries and territories to gauge the success of various strategies that have been implemented to target specific NCDs or their risk factors.

Our study had various limitations that could result in either an under- or overestimation of the presented mortality rates. One limitation was that the denominator data were not available for the entirety of the study years from any of the countries. A second was that trends could not be analyzed for a few of the countries due to either a small number of deaths annually or missing years of data. For example, one country with a small population reported more than 10% “unknown/missing/invalid” cause of death codes, which could compromise the ability to accurately rank the cancer site. A third limitation was that deaths due to diabetes are a problematic index of the level of diabetes care because diabetes is frequently underreported as a cause of death on death certificates. Finally, we were unable to account for the variability that exists among countries regarding the inclusion/exclusion of nonresident deaths in mortality statistics. In addition, the results should be interpreted with caution, given that age-standardized mortality rates may be unreliable in some countries due to small numbers.

Improvements in the collection of health and vital statistics data, along with improvements in systems for surveillance of NCDs, are needed to increase reporting accuracy of regional data for future investigations in the Caribbean.

Despite the various limitations, our results comprehensively highlight country-specific mortality rates and trends for deaths due to NCDs. These findings could be used to inform policy and interventions for strengthening NCD prevention and control in the 2 United States territories and the 20 other English- or Dutch-speaking Caribbean countries and territories that we studied.

The WHO NCD global monitoring framework has nine global NCD targets for 2025, including the reduction of premature mortality from NCDs by 25% (26). Future work to determine whether Caribbean countries and territories are on track to meet this target could be extremely beneficial to the region.

All the authors conceived the original idea, guided that analysis, interpreted the results, drafted parts, and reviewed the paper. All the authors reviewed and approved the final version.

The authors hold sole responsibility for the views expressed in the manuscript, which may not necessarily reflect the opinion or policy of the RPSP/PAJPH or PAHO. In addition, the findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the U.S. Centers for Disease Control and Prevention.