Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jul 1.
Published in final edited form as: Prev Med. 2014 Mar 25;64:41–47. doi: 10.1016/j.ypmed.2014.03.018

Socioeconomic inequalities in premature mortality in Colombia, 1998-2007: The double burden of non-communicable diseases and injuries

Ivan Arroyave 1, Alex Burdorf 1, Doris Cardona 2, Mauricio Avendano 1,2,3
PMCID: PMC4067972  NIHMSID: NIHMS591706  PMID: 24674854

Abstract

Objectives

Non-communicable diseases have become the leading cause of death in middle-income countries, but mortality from injuries and infections remains high. We examined the contribution of specific causes to disparities in adult premature mortality (ages 25-64) by educational level from 1998 to 2007 in Colombia.

Methods

Data from mortality registries were linked to population censuses to obtain mortality rates by educational attainment. We used Poisson regression to model trends in mortality by educational attainment and estimated the contribution of specific causes to the Slope Index of Inequality.

Results

Men and women with only primary education had higher premature mortality than men and women with post-secondary education (RRmen=2·60, 95% confidence interval [CI]:2·56, 2·64; RRwomen=2·36, CI:2·31, 2·42). Mortality declined in all educational groups, but declines were significantly larger for higher-educated men and women. Homicide explained 55·1% of male inequalities while non-communicable diseases explained 62·5% of female inequalities and 27·1% of male inequalities. Infections explained a small proportion of inequalities in mortality.

Conclusion

Injuries and non-communicable diseases contribute considerably to disparities in premature mortality in Colombia. Multi-sector policies to reduce both interpersonal violence and non-communicable disease risk factors are required to curb mortality disparities.

Keywords: Mortality Determinants, Colombia, Educational Status, Burden of Illness, Health Transition, Health Status Disparities, Socioeconomic Factors, Age-Specific Death Rate

INTRODUCTION

In most high-income countries, approximately two thirds of socioeconomic inequalities in mortality are attributable to cardiovascular disease and cancer, with less than 5% attributable to injuries and communicable diseases (Huisman et al., 2005). This pattern may be markedly different in low- and middle-income countries, where non-communicable diseases have become a leading cause of death, but mortality from both communicable diseases and injuries remains relatively high (Frenk et al., 1991). While lower socioeconomic status is often associated with higher mortality from ‘poverty-related diseases’ such as preventable infections (Singh and Singh, 2008), it is less clear how socioeconomic status might relate to conditions associated with modern lifestyles such as cardiovascular disease (Singh and Singh, 2008). The contribution of different causes to socioeconomic inequalities in mortality has been documented in wealthy nations (Fawcett et al., 2005; Huisman et al., 2005; Kunst et al., 1998b; Wong et al., 2002), while few studies have focused on low- and middle-income countries (Belon et al., 2012).

Colombia faces relatively high mortality from communicable diseases and injuries, as well as high mortality from non-communicable diseases (Mayorga, 2004). Rates of premature mortality from non-communicable diseases are comparable to those in high-income countries, while mortality from infections and injuries are four times higher (Appendix Figure 1) (World Health Organization, 2012). This pattern has resulted in a double burden, with injuries and communicable diseases accounting for approximately half of all deaths, and non-communicable diseases for another half (Mayorga, 2004; World Health Organization, 2012). A potential hypothesis is that the increasing burden of non-communicable disease mortality (Mayorga, 2004) has disproportionately affected the lower socioeconomic groups, which also have higher mortality from infectious diseases and injuries (Mayorga, 2004).

Classified as a middle-high income country (World Bank, 2011), Colombia has experienced improvements in socioeconomic and healthcare indicators over the last decades. Between 1998 and 2007, constant GDP per capita grew on average by 1.9% per year. The percentage of population living in poverty (less than US$2 per day) declined from 14.1% in 1998 to 7.5% in 2007 (World Bank, 2011), and healthcare insurance coverage increased from 59.8% to 92.5% (Arroyave et al., 2013). Educational attainment has also risen (Appendix-Figure 2) , with noticeable increases in the proportion of population with secondary and tertiary education (IIASA/VID, 2010). Despite these improvements, inequalities in Colombia remain high by international standards; In 1999-2003, the Gini coefficient of income inequality was 55.9% (World Bank, 2011).

In this study, we examine trends in socioeconomic inequalities in mortality and estimate the contribution of specific causes of death to these differentials between 1998 and 2007 in Colombia. We hypothesized that mortality from non-communicable diseases, infections and injuries contribute each to socioeconomic differences in mortality. However, we expected an increasing concentration of non-communicable diseases in the lower socioeconomic groups, leading to widening socioeconomic inequalities in total mortality.

METHODS

Data

Data were obtained from official registries of the National Administrative Department of Statistics, which collects and harmonizes data on deaths based on international guidelines. For all deceased individuals (633,906 deaths), data were collected on sex, age of death and educational level. Causes of death were coded according to the International Classification of Diseases (ICD-10) and aggregated into 10 major causes of death grouped into four broad categories: non-communicable diseases, injuries, infections and other causes. Table-Appendix 1 shows specific ICD-10 codes for each cause of death.

Data on population counts were obtained based on the following procedure: First, we estimated the distribution of education by 5-year age group, sex and year, based on data from census and national surveys harmonized by the International Institute of Applied System Analysis (IIASA) and the Vienna Institute of Demography (VID) as part of the IIASA/VID database (IIASA/VID, 2010). We then combined this information with data on national population counts from the National Statistics Office to obtain the number of population by educational attainment. IIASA/VID Data were only available per quinquennium; therefore, we performed demographic projections to obtain population counts for in-between years using the demographic Software PASEX (U.S. Census Bureau, 2011).

Data on educational level was missing for approximately a third of deceased cases. We used multiple imputation methods implemented in SAS through the IMPUTE procedure to impute educational level for these cases. This was done to avoid bias due to the potentially higher rates of missing education for lower educated individuals, and to minimize the potential for numerator/denominator bias (Kunst et al., 1998a). This procedure fits a sequence of regression models and draws values from the corresponding predictive distributions. The sequential regression procedure was applied based on a model that included sex, region, age and marital status as covariates. Imputation was not possible in 6.8% of cases. Full details on the procedure are available elsewhere (Raghunathan et al., 2001).

We distinguished three groups based on highest educational level attained: (a) completed primary school, (b) completed secondary school, and (c) completed tertiary education. We excluded individuals below age 25, because many would not have completed their education before this age. In addition, we focused on adult premature mortality (mortality below age 65), an indicator of population health believed to be strongly influenced by social, economic and environmental factors (World Health Organization, 2008), and a common indicator of health system performance (Smith et al., 2009). While some premature deaths are unavoidable, a substantial part of premature mortality is avoidable through public health programmes and policies that address the social determinants of health (World Health Organization, 2008).

Methods of analysis

We first calculated age-standardized mortality rates by educational level, sex and year. Rates were standardized using the direct method based on the WHO standard population of 1997, which better reflects the age structure of the world population than the Segi standard population (Ahmad et al., 2001). Subsequently, we implemented separate Poisson regression models with deaths as dependent variable and the natural log of person-years as offset variable, incorporating age and educational level as independent variables.

To assess mortality trends by educational level, we estimated the annual percent change in mortality (APC) based on a Poisson model that incorporated an interaction between educational level and year. The APC measures the average rate of change in the mortality rate per year (Clegg et al., 2009). At a second stage, we estimated rate ratios (RR) of mortality by educational level. To assess changes in inequalities ‘controlling’ for changes in the educational distribution, we estimated the slope index of inequality (SII) and the relative index of inequality (RII). To construct these measures, educational groups are first ordered from lowest to highest. The population in each educational level covers a rage in the cumulative distribution of the population. Mortality is then regressed on the mid-point of the cumulative distribution of education for each group (Mackenbach and Kunst, 1997). The RII can be interpreted as the ratio of mortality between a hypothetical person whose relative rank in the distribution of education is zero and a person whose relative rank in the cumulative distribution of education is 100% (Mackenbach and Kunst, 1997). The SII corresponds to the equivalent absolute rate difference between these two points. Further details on the RII and SII are available elsewhere (Mackenbach and Kunst, 1997).

Regression analyses were conducted in each of the five multiple databases generated by the multiple imputation process. Since results were nearly identical for all imputations, we used the PROC MIANALYZE procedure in SAS to combine estimates from all databases and adjust standard errors, accounting for the uncertainty in the imputation (SAS Institute, 2008). This procedure reads the parameter estimates and associated covariance matrix for each imputed dataset, and then derives valid multivariate inferences for these parameters. This allows for valid statistical inference that appropriately reflects uncertainty due to missing values (SAS Institute, 2008). All analyses were conducted in SAS® version 9.2.

RESULTS

Table 1 shows mortality rates at ages 25-64 years between 1998 and 2007 in Colombia. 633,905 deaths occurred from 1998 to 2007, with male deaths accounting for two thirds of overall deaths (66·0%), mainly owing to exceptionally high rates of homicide mortality. Non-communicable diseases accounted for half of all female mortality (50·4%) while 46·3% of mortality among men were due to injuries. Infections accounted for around 7% of deaths.

Table 1.

Premature mortality rates per 100,000 person-years by educational level, ages 25-64, 1998-2007, Colombia

Deaths Percentage of
deaths		 Standardized mortality rates
per 100,000 person-years
Men
 Women
 Men
 Women
 Men
 Women
DEATHS BY CAUSE
 Cardiovascular diseases (CVD) 70,757 51,976 18.1% 25.8% 44.0 30.5
 Malignant neoplasm 49,809 65,601 12.7% 32.5% 37.1 47.9
 Diabetes mellitus (DM) 9,769 10,616 2.5% 5.3% 5.6 5.0
 Chronic lower respiratory diseases (CLRD) 7,298 5,729 1.9% 2.8% 3.0 2.4
  Total Non-Communicable diseases 137,633 133,922 35.2% 66.4% 89.7 85.8
 Traffic Accident 29,104 6,373 7.4% 3.2% 33.1 6.8
 Suicide 9,399 1,929 2.4% 1.0% 10.5 1.9
 Homicide 121,983 9,966 31.2% 4.9% 124.3 10.0
 Other injuries 25,056 4,057 6.4% 2.0% 28.7 4.3
  Total Injuries 185,542 22,325 47.4% 11.1% 196.6 23.0
 Tuberculosis (TB) and sequelae 4,200 1,861 1.1% 0.9% 4.0 1.9
 HIV disease (AIDS) 14,369 3,218 3.7% 1.6% 15.4 3.1
 Pneumonia 5,057 3,494 1.3% 1.7% 4.9 3.0
 Other infectious diseases 4,894 3,520 1.3% 1.7% 5.1 3.3
  Total Infectious Diseases 28,520 12,093 7.3% 6.0% 29.5 11.2
 Other Non-Communicable Diseases 24,225 17,281 6.2% 8.6% 21.5 14.7
 Rest of diseases 9,602 13,309 2.5% 6.6% 9.8 14.1
 Ill Defined Causes 5,841 2,877 1.5% 1.4% 6.4 2.8
  Total Other Diseases 39,668 33,467 10.1% 16.6% 37.7 31.6
DEATHS BY EDUCATIONAL ATTAINMENT
  Primary 244,971 139,358 62.6% 69.1% 439.9 192.2
  Secondary 121,136 52,386 31.0% 26.0% 312.2 126.6
  Tertiary 25,258 10,064 6.5% 5.0% 166.0 81.8
  TOTAL DEATHS 391,363 201,807 100% 100% 353.5 151.7
OPULATION Person-years Percentage of
person-years
Men
Women
Men
Women
  Primary 40,773,078 42,981,451 46.7% 45.9%
  Secondary 33,768,114 37,506,347 38.6% 40.1%
  Tertiary 12,849,341 13,079,674 14.7% 14.0%
    TOTAL: 87,390,533 93,567,472 100% 100%
Open in a new tab
*

Percentage of deaths out of total mortality separtely for men and women

**

Percentage distribution of educational attainment out of total population sperately for men and women

Figure 1 shows premature mortality rates by educational level. Men and women with lower levels of education had higher premature mortality from any cause of death than their higher-educated counterparts. Rate ratios summarising differences in mortality across educational groups are presented in Table 2. Less-educated men and women had higher rates of mortality for all causes than their higher-educated counterparts (RRmen=2·60, 95% confidence interval [CI]:2·56-2·64; RRwomen=2·36, 95%CI:2·31-2·42). Inequalities were largest for injuries among men (RR=3·64, 95%CI:3·54-3·3), while among women, they were largest for infections (RR =4·22, 95%CI:3·83-4·65), particularly for Tuberculosis and HIV/AIDS.

Figure 1.

Figure 1

Open in a new tab

Age-standardized premature mortality rates per 100,000 population by educational level, ages 25-64, 1998-2007, Colombia

Table 2.

Rate Ratios (RR) of age-standardized premature mortality rates by educational level, ages 25-64, 1998-2007, Colombia

MEN
 WOMEN
 MEN
 WOMEN
 MEN
 WOMEN
 MEN
 WOMEN
Cardiovascular diseases (CVD)
 Traffic Accident
 Tuberculosis (TB) and sequelae
 Other Non-Communicable Diseases (ONCD)
Primary 1.88 ( 1.82 , 1.95 ) 2.81 ( 2.68 , 2.95 ) 2.28 ( 2.17 , 2.40 ) 1.52 ( 1.32 , 1.74 ) 4.70 ( 3.87 , 5.70 ) 6.60 ( 4.85 , 8.99 ) 2.24 ( 2.11 , 2.38 ) 2.67 ( 2.42 , 2.94 )
Secondary 1.54 ( 1.48 , 1.59 ) 1.67 ( 1.59 , 1.77 ) 1.91 ( 1.82 , 2.01 ) 1.26 ( 1.12 , 1.41 ) 2.49 ( 1.96 , 3.16 ) 2.80 ( 1.99 , 3.92 ) 1.54 ( 1.45 , 1.64 ) 1.61 ( 1.44 , 1.79 )
Tertiary (Ref) 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Malignant neoplasm
Suicide
HIV disease (AIDS)
Rest of diseases
Primary 1.54 ( 1.49 , 1.60 ) 1.58 ( 1.52 , 1.63 ) 2.73 ( 2.51 , 2.98 ) 3.45 ( 2.84 , 4.19 ) 1.55 ( 1.44 , 1.67 ) 5.50 ( 4.51 , 6.72 ) 2.61 ( 2.38 , 2.86 ) 3.18 ( 2.93 , 3.47 )
Secondary 1.35 ( 1.30 , 1.41 ) 1.31 ( 1.26 , 1.35 ) 1.84 ( 1.69 , 2.00 ) 1.89 ( 1.53 , 2.34 ) 1.90 ( 1.77 , 2.03 ) 3.36 ( 2.78 , 4.07 ) 1.67 ( 1.52 , 1.83 ) 1.94 ( 1.78 , 2.12 )
Tertiary (Ref) 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Diabetes mellitus
Homicide
Pneumonia
Ill Defined Causes
Primary 1.72 ( 1.57 , 1.90 ) 3.33 ( 2.90 , 3.83 ) 4.22 ( 4.10 , 4.34 ) 3.57 ( 3.27 , 3.90 ) 2.76 ( 2.39 , 3.18 ) 3.62 ( 2.98 , 4.39 ) 2.27 ( 2.00 , 2.58 ) 2.93 ( 2.22 , 3.87 )
Secondary 1.71 ( 1.55 , 1.89 ) 1.78 ( 1.54 , 2.05 ) 2.57 ( 2.49 , 2.65 ) 2.04 ( 1.87 , 2.23 ) 1.80 ( 1.58 , 2.06 ) 2.06 ( 1.67 , 2.54 ) 1.59 ( 1.42 , 1.78 ) 1.47 ( 1.07 , 2.01 )
Tertiary (Ref) 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Chronic lower respiratory diseases (CLRD)
Other injuries
Other infectious diseases
OTHER DISEASES
Primary 2.98 ( 2.57 , 3.47 ) 3.48 ( 2.94 , 4.11 ) 3.66 ( 3.35 , 4.01 ) 2.93 ( 2.50 , 3.43 ) 2.48 ( 2.14 , 2.87 ) 2.96 ( 2.50 , 3.50 ) 2.33 ( 2.23 , 2.43 ) 2.91 ( 2.74 , 3.08 )
Secondary 1.89 ( 1.60 , 2.22 ) 1.63 ( 1.36 , 1.95 ) 2.07 ( 1.87 , 2.28 ) 1.48 ( 1.24 , 1.76 ) 1.60 ( 1.38 , 1.86 ) 1.80 ( 1.50 , 2.14 ) 1.58 ( 1.51 , 1.65 ) 1.74 ( 1.63 , 1.84 )
Tertiary (Ref) 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
NON-COMMUNICABLE DISEASES
INJURIES
INFECTIOUS DISEASES
ALL DEATHS
Primary 1.77 ( 1.73 , 1.81 ) 2.06 ( 2.01 , 2.12 ) 3.64 ( 3.54 , 3.73 ) 2.64 ( 2.45 , 2.85 ) 2.14 ( 2.02 , 2.26 ) 4.22 ( 3.83 , 4.65 ) 2.60 ( 2.56 , 2.64 ) 2.36 ( 2.31 , 2.42 )
Secondary 1.48 ( 1.45 , 1.52 ) 1.43 ( 1.39 , 1.47 ) 2.32 ( 2.25 , 2.38 ) 1.63 ( 1.53 , 1.75 ) 1.89 ( 1.79 , 1.99 ) 2.40 ( 2.17 , 2.65 ) 1.87 ( 1.83 , 1.90 ) 1.56 ( 1.52 , 1.60 )
Tertiary (Ref) 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Open in a new tab

Appendix-Figure 3 shows that premature mortality declined among both men and women over the study period. However, mortality rates for those with primary and secondary education remained relatively constant or grew for deaths from infectious disease and other causes, while rates for higher educated men steadily declined for all causes. While mortality from injuries declined steadily among men with middle and higher education, it increased during the first years for lower-educated men, and only started to decline in 2002. Figure 2 summarizes trends in mortality differences by education on the basis of the RII. For both men and women, inequalities in total and cause-specific mortality widened over the first half of the period (1998-2003), but remained stable over the second half (2004-2007).

Figure 2.

Figure 2

Open in a new tab

Trends of the Relative Index of Inequality (RII) of age age-standardized premature mortality rates by educational level, ages 25 rates by educational level, ages 25-64, 1998-2007, Colombia

APC estimates in Figure 3 show that the initial increase in inequalities by educational level is due to more favourable trends among the higher-educated groups. Among men, mortality declined by 4·5% (95%CI: −5·0%, −4·0%) per year in men with tertiary education, as compared to 2·3% (95%CI: −2·6%, −2·1%) in men with secondary education and 1·5% (95%CI: −1·7%, −1·4%) in men with primary education or less. Similar results were observed among women, although differences in the APC were not significant. The largest difference in trends was for injuries among men and women. Similar trends were observed for non-communicable disease mortality among men, while declines were similar for women from all educational groups. Although confidence intervals were wide, results suggest that those with primary education experienced an increase in infectious disease mortality, while those with tertiary education experienced no change.

Figure 3.

Figure 3

Open in a new tab

Annual percentage change (APC) of age-standardized premature mortality rates by educational level, ages 25 educational level, ages 25-64, 1998-2007, Colombia

Figure 4 shows the contribution of each cause of death to absolute differences in premature mortality by education measured with the SII. Absolute differences in mortality were nearly twice larger for men (SII=402·4 deaths for 100,000 population) than for women (SII=228·9). This difference was almost entirely due to the large contribution of injuries, particularly homicide, to inequalities in mortality among men, which overall explained 55·1% of male inequalities. Non-communicable disease mortality was the second largest contributor among men, accounting for 27·1% of inequalities in total mortality. Among women, non-communicable diseases were by far the largest contributor to inequalities, explaining 62·5% of inequalities in total mortality. Infections explained only 5·9% of differences in mortality by education among men and 8·0% among women.

Figure 4.

Figure 4

Open in a new tab

Slope Index of Inequality (SII) of age age-standardized premature mortality rates per 100,000 person-years by educational level, ages 25 years by educational level, ages 25-64, 1998-2007, Colombia

DISCUSSION

Inequalities in premature mortality by education in Colombia widened over the first half of the study period and remained constant over the second half. Mortality from injuries, particularly homicide, explains more than half of inequalities among men, while non-communicable diseases are the most important contributor to female inequalities and the second contributor among men. Infections explain a relatively small proportion of inequalities in premature mortality. Our results highlight the need for a shift in focus towards policies addressing the increasing contribution of non-communicable disease and injuries to socioeconomic inequalities in premature mortality.

Explanation of results

Our study suggests that lower-educated men and women have benefited significantly less from declining premature mortality than their higher-educated counterparts. Several explanations could account for this pattern, including inequalities by educational level in social, economic and working conditions; access to health care; and risk factor prevalence. Our decomposition by cause of death sheds some light on the role of some of these mechanisms.

The most striking finding from our study is the large contribution of homicide to socioeconomic inequalities in premature mortality among men. Homicide rates in Colombia have declined (Acero-Álvarez, 2011; Bonilla Mejía, 2010) but remain among the highest worldwide (World Health Organization, 2012). We found that homicide is primarily concentrated among lower-educated men, and it is disproportionately high for young men (Acero-Álvarez, 2011). Lower-educated men face high levels of poverty, unemployment, social disruption and risky behaviours (e.g., alcohol, drug use, smoking), and are more likely to live in deprived areas (Acero-Álvarez, 2011). Colombia has one of the highest levels of income inequality in Latin America (UNDP, 2010), which may contribute to high youth homicide rates (Briceño-León et al., 2008). Our findings underscore the significance of homicide as a major contributor to male mortality inequalities.

Our study also suggests that traffic accidents have large socioeconomic gradients and contribute importantly to socioeconomic differences in mortality in Colombia. Lower education has been linked to higher reliance on unsafe forms of transportation (Males, 2009). Vehicle safety infrastructure is less well-developed in socially deprived areas, where individuals may be less likely to comply with safety regulations on seat belt use, driving while drinking, and speed limit enforcement (Males, 2009; Rodríguez et al., 2003).

Mirroring findings for high-income countries (Fawcett et al., 2005; Huisman et al., 2005; Kunst et al., 1998b; Wong et al., 2002), we found that non-communicable diseases are a leading contributor to inequalities in mortality by educational level in Colombia. Socioeconomic inequalities in non-communicable diseases have been associated with the unequal distribution of behavioural risk factors, particularly smoking, alcohol consumption, an unhealthy diet and a sedentary lifestyle (Adler and Newman, 2002). Existing evidence suggests that, as in high income countries, lower education is associated with a worse risk factor profile in Colombia. Data from 2007 suggests that the prevalence of smoking was 41% among Colombians with primary education or less, as opposed to 26% among those with a college education (Storr et al., 2008). Similarly, 26% of lower-educated Colombians aged 25-50 years have at least a risk factor for cardiovascular disease, as opposed to only 5·9% in those with a University degree (Patiño-Villada et al., 2011). Trends in infections, on the other hand, might reflect socioeconomic differences in both preventive and curative care, which may remain despite large increases in health insurance coverage (Arroyave et al., 2013; Gaviria et al., 2006). Noticeably, there are large inequalities in the availability of running water, sewage systems and adequate housing in Colombia (UNDP, 2010), which may be more important in generating inequalities in communicable diseases.

Limitations of the study

Despite several strengths, some limitations should be considered. Data on mortality were obtained from mortality registries, while data on population counts came from censuses. This may have led to the so-called numerator/denominator bias, which generally results in an overestimation of inequalities (Kunst et al., 1998a). For some years, data on population size were obtained from demographic projections, as census were conducted in 1985, 1993 and 2005 (DANE, 2012). We focused on premature mortality, given the public health relevance of this measure and strong association with social determinants and health system performance (Smith et al., 2009; World Health Organization, 2008). Further research is required to examine mortality patterns for older ages.

As in other middle-income countries (Murray and Lopez, 1997) underregistration of deaths remains a problem in Colombia, particularly in the poorest regions (Rodríguez-García, 2007). For example, the estimated proportion of registered deaths in the Choco region, one of the poorest in the country, was only 25%, compared to 90% or higher in most other regions (Rodríguez-García, 2007). Our estimates of inequalities are likely an underestimation, because those with lower education are more likely to live in areas with higher underregistration. We may also have underestimated the increase in inequalities, because underregistration has decreased over the study period (Florez and Méndez, 1995; Rodríguez-García, 2007). Our results, therefore, are indicative of potentially larger inequalities in mortality by education.

It is likely that part of the differences in mortality by education observed in our study reflects regional differences in mortality. Unfortunately, no data are available on mortality by educational level separately by region. To partly address this question, however, maps in Appendix Figure 4 show age-standardized mortality rates and average years of schooling for each region in 2002 (similar regional patterns are observed for other years). Based on these aggregate data, we find only a weak correlation between regional average years of schooling and mortality (r=0.07, p=0.73). Nevertheless, more detailed data is required to fully examine to what extent regional variations explain differences in mortality by education.

Information on education was missing for 34·2% of death records. This may have led to an underestimation of inequalities, as missing values are likely to be more common for the least educated (Rodríguez-García, 2007). We imputed missing values on education based on a rich set of variables available for most deceased individuals, partly minimizing potential bias.

Conclusion

Mortality from both injuries and non-communicable diseases contribute considerably to disparities in premature mortality in Colombia. The striking contribution of homicide to socioeconomic differences in mortality among men highlights the need for public policies that address the profound social and economic factors that underlie interpersonal violence in Colombia. At the same time, the increasing contribution of non-communicable diseases calls for urgent prevention policies for curbing the increasing prevalence of non-communicable disease risk factors in the lower socioeconomic groups.

Supplementary Material

01

Research highlights.

  • - Both injuries and non-communicable diseases contribute to mortality inequalities

  • - Homicide contributed more than half of inequalities in male premature mortality

  • - Policies on violence and chronic disease risk factors are key to curb disparities

Acknowledgments

Funding disclosure I. Arroyave was supported by the European Union Erasmus Mundus Partnerships Programme Erasmus-Columbus and the Programme Enlazamundos), and the Direction of Research of the Universidad CES, Medellin-Colombia (grant No 2012DI09). Dr. Mauricio Avendano was supported by a Starting Researcher grant from the European Research Council (ERC grant No 263684), the National Institute on Aging (grants R01AG040248 and R01AG037398), a fellowship from the Erasmus University, and the McArthur Foundation Research Network on Ageing.

Role of funding source The sponsors of the authors had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Authors’ contributions I. Arroyave was the leading author and developed the article idea, constructed and analysed the data set, and wrote drafts of article. A. Burdorf contributed to interpretation of results and commented on all drafts. D. Cardona contributed to the quantitative analysis and commented on all drafts of the article. M. Avendano analysed data, wrote sections of the article, and contributed to the coordination of all steps of the analysis and article preparation.

Ethics committee approval This article is based on secondary analysis of data on deaths and population counts in aggregate form made publically available by the National Statistics Office in Colombia. Ethical approval for this study was not required.

Conflict of interest statements We are pleased to report no conflict of interest by any of the authors of this paper. There is no financial support for this work that could have influenced its outcome.

REFERENCES

  1. Acero-Álvarez A.d.P. Descripción del Comportamiento del Homicidio. Colombia, 2010. In: Forensis, editor. Datos oficiales sobre la violencia en Colombia en el 2010. Instituto Nacional de Medicina Legal y Ciencias Forenses; Bogota: 2011. [Google Scholar]
  2. Adler NE, Newman K. Socioeconomic Disparities In Health: Pathways And Policies. Health Affairs. 2002;21:60–76. doi: 10.1377/hlthaff.21.2.60. [DOI] [PubMed] [Google Scholar]
  3. Ahmad OB, Boschi-Pinto C, Lopez AD, Murray CJ, Lozano R, Inoue M. Age standardization of rates: A new WHO standard, GPE Discussion Paper Series. World Health Organization; Geneva: 2001. [Google Scholar]
  4. Arroyave I, Cardona D, Burdorf A, Avendaño M. The Impact of Increasing Health Insurance Coverage on Disparities in Mortality: Health Care Reform in Colombia, 1998–2007. American Journal of Public Health. 2013:e1–e7. doi: 10.2105/AJPH.2012.301143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Belon AP, Barros MB, Marin-Leon L. Mortality among adults: gender and socioeconomic differences in a Brazilian city. BMC public health. 2012;12:39. doi: 10.1186/1471-2458-12-39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bonilla Mejía L. Demografía, juventud y homicidios en Colombia, 1979-2006. Lecturas de Economía. 2010:77–102. [Google Scholar]
  7. Briceño-León R, Villaveces A, Concha-Eastman A. Understanding the uneven distribution of the incidence of homicide in Latin America. International Journal of Epidemiology. 2008;37:751–57. doi: 10.1093/ije/dyn153. [DOI] [PubMed] [Google Scholar]
  8. Clegg LX, Hankey BF, Tiwari R, Feuer EJ, Edwards BK. Estimating average annual per cent change in trend analysis. Statistics in Medicine. 2009;28:3670–82. doi: 10.1002/sim.3733. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. DANE . Estadisticas Vitales, Informacion DANE. Departamento Administrativo Nacional de Estadistica; Bogota: 2012. p. Bases de datos economicas y demograficas. [Google Scholar]
  10. Fawcett J, Blakely T, Kunst A. Are Mortality Differences and Trends by Education Any Better or Worse in New Zealand? A Comparison Study with Norway, Denmark and Finland, 1980–1990s. European Journal of Epidemiology. 2005;20:683–91. doi: 10.1007/s10654-005-7923-y. [DOI] [PubMed] [Google Scholar]
  11. Florez CE, Méndez R. El Nivel del Subregistro de las Defunciones: Colombia 1990. CEDE; Colombia: 1995. pp. 69–85. [Google Scholar]
  12. Frenk J, Frejka T, Bobadilla JL, Stern C, Lozano R, Sepúlveda J, José M. La transición epidemiológica en América Latina, Conferencia Internacional de Población. Boletín de la Oficina Sanitaria Panamericana (OSP); Nueva Delhi: 1991. pp. 485–96. [PubMed] [Google Scholar]
  13. Gaviria A, Medina C, Mejia C. Assessing Health Reform in Colombia: From Theory to Practice. Journal of LACEA Economia. 2006;7:29–72. [Google Scholar]
  14. Huisman M, Kunst AE, Bopp M, Borgan J-K, Borrell C, Costa G, Deboosere P, Gadeyne S, Glickman M, et al. Educational inequalities in cause-specific mortality in middle-aged and older men and women in eight western European populations. The Lancet. 2005;365:493–500. doi: 10.1016/S0140-6736(05)17867-2. [DOI] [PubMed] [Google Scholar]
  15. IIASA/VID . IVEP-1.0 (IIASA/VID education database) International Institute of Applied System Analysis (IIASA); Vienna Institute of Demography of the Austrian Academy of Sciences (VID); Viena: 2010. [Google Scholar]
  16. Kunst AE, Groenhof F, Borgan J, Costa G, Desplanques G, Faggiano F, Hemström O, Martikainen P, Vågerö D, et al. Socio-economic inequalities in mortality. Methodological problems illustrated with three examples from Europe. Revue d’epidemiologie et de sante publique. 1998a;46:467–79. [PubMed] [Google Scholar]
  17. Kunst AE, Leon DA, Groenhof F, Mackenbach JP. Occupational class and cause specific mortality in middle aged men in 11 European countries: comparison of population based studies. BMJ. 1998b;316:1636–42. doi: 10.1136/bmj.316.7145.1636. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Mackenbach JP, Kunst AE. Measuring the magnitude of socio-economic inequalities in health: An overview of available measures illustrated with two examples from Europe. Social Science & Medicine. 1997;44:757–71. doi: 10.1016/s0277-9536(96)00073-1. [DOI] [PubMed] [Google Scholar]
  19. Males MA. Poverty as a determinant of young drivers’ fatal crash risks. Journal of Safety Research. 2009;40:443–48. doi: 10.1016/j.jsr.2009.10.001. [DOI] [PubMed] [Google Scholar]
  20. Mayorga C. Tendencia de la mortalidad y sus determinantes como parte de la transición epidemiológica en Colombia. Revista Gerencia y Políticas de Salud. 2004 [Google Scholar]
  21. Murray CJL, Lopez AD. Mortality by cause for eight regions of the world: Global Burden of Disease Study. The Lancet. 1997;349:1269–76. doi: 10.1016/S0140-6736(96)07493-4. [DOI] [PubMed] [Google Scholar]
  22. Patiño-Villada FA, Arango-Vélez EF, Quintero-Velásquez MA, Arenas-Sosa MM. Factores de riesgo cardiovascular en una población urbana de Colombia. Revista de Salud Pública. 2011;13:433–45. [PubMed] [Google Scholar]
  23. Raghunathan T, Lepkowski J, Van Hoewyk J, Solenberger P. A Multivariate Technique for Multiply Imputing Missing Values Using a Sequence of Regression Models. Survey Methodology. 2001;27:85–95. [Google Scholar]
  24. Rodríguez-García J. Desigualdades socioeconómicas entre departmentos y su asociación con indicadores de mortalidad en Colombia en 2000. Revista Panamericana de Salud Pública. 2007;21:111–24. doi: 10.1590/s1020-49892007000200006. [DOI] [PubMed] [Google Scholar]
  25. Rodríguez DY, Fernández FJ, Velásquez HA. Road traffic injuries in Colombia. Injury Control and Safety Promotion. 2003;10:29–35. doi: 10.1076/icsp.10.1.29.14119. [DOI] [PubMed] [Google Scholar]
  26. SAS Institute, I. The MIANALYZE Procedure. In: Inc., S.I., editor. User’s Guide SAS/STAT® 9.2. Version 8 ed SAS Institute Inc.; Cary, NC: 2008. pp. 201–33. [Google Scholar]
  27. Singh AR, Singh SA. Diseases of poverty and lifestyle, well-being and human development. Mens Sana Monogr. 2008;6:187–225. doi: 10.4103/0973-1229.40567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Smith P, Mossialos E, Papanicolas I, Leatherman S. Performance Measurement for Health System Improvement: Experiences, Challenges and Prospects. Cambridge University Press; Cambridge: 2009. [Google Scholar]
  29. Storr CL, Cheng H, Posada-Villa J, Aguilar-Gaxiola S, Anthony JC. Adult smokers in Colombia: who isn’t giving it up? Addict Behav. 2008;33:412–21. doi: 10.1016/j.addbeh.2007.10.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. U.S. Census Bureau . Population Analysis System (PASEX), Population Analysis Spreadsheets. 2.04g ed Washington DC: 2011. [Google Scholar]
  31. UNDP . Acting on the future: breaking the intergenerational transmission of inequality, Regional Human Development Report for Latin America and the Caribbean. 1 ed United Nations Development Programme (UNDP); San José, C.R.: 2010. p. 208. [Google Scholar]
  32. Wong MD, Shapiro MF, Boscardin WJ, Ettner SL. Contribution of Major Diseases to Disparities in Mortality. New England Journal of Medicine. 2002;347:1585–92. doi: 10.1056/NEJMsa012979. [DOI] [PubMed] [Google Scholar]
  33. World Bank . World Development Indicators (WDI) World Bank Data (WBD); 2011. [Google Scholar]
  34. World Bank World Health Organization . Final Report of the Commission on Social Determinants of Health. World Health Organization; Geneve: 2008. Closing the gap in a generation: Health equity through action on the social determinants of health. Final Report of the Commission on Social Determinants of Health. [Google Scholar]
  35. World Health Organization . Disease and injury regional estimates: Cause-specific mortality: regional estimates for 2008. World Health Organization; 2012. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS591706-supplement-01.pdf (3.5MB, pdf)

RESOURCES