You are here

Article Text

Article menu

Download PDFPDF

Original article
Association between economic growth and injury mortality among seniors in Colombia
  1. Antonio J Trujillo1,
  2. Adnan A Hyder1,2,
  3. Fernando Ruiz3
  1. 1Health Systems Program, Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
  2. 2International Injury Research Unit, Department of International Health and Center for Injury Research & Policy, Johns Hopkins University, Baltimore, Maryland, USA
  3. 3Cendex, Pontificia, Universidad Javeriana, Colombia
  1. Correspondence to Dr Antonio J Trujillo, Department of International Health, Johns Hopkins Bloomberg School of Public Health, 615 North Wolfe Street, Suite E-8132, Baltimore, MD 21205, USA; atrujill{at}jhsph.edu

Abstract

Background Injuries among seniors are recognised as an important public health problem not only in developed countries but also in middle-income countries. There is ample epidemiological literature that relates economic growth to the reduction of infectious and childhood diseases. Less evidence exists to document if economic growth alone is enough to reverse the increasing trends of injury mortality and morbidity among seniors in a middle-income country.

Aim To investigate the association between economic growth and injury deaths among older people in Colombia.

Method Using data from Colombia, 1979–2006 (n=28), time-series models were used to ascertain if the variation over time in injury mortality among seniors is related to short-term oscillations in economic performance. Four empirical specifications usually used in the analysis of such data were implemented. Models were run by type of injury and gender.

Results A negative but moderate effect of economic growth was found on injury deaths among older people. The reported elasticity was between −0.98 and −1.26. Men benefit from economic growth more than women. Economic growth seems to reduce traffic injuries, suicides and homicides. A positive association was also found between falls and growth in gross domestic product.

Conclusions The results indicate a non-homogeneous association between economic growth and injury deaths among seniors in Colombia. This association is usually stronger in a negative direction among children and younger adults. Although more research is needed to understand the causal relationship between economic growth and injury, the association found may suggest that economic growth may not be sufficient to reverse injury deaths among older people; therefore, additional health policies need to be in place to reduce mortality due to preventable injuries in seniors.

  • Economic growth
  • unintentional Injury
  • mortality
  • elderly
  • Colombia
  • developing nations
  • elderly
  • gender

https://doi.org/10.1136/ip.2009.023176

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Current estimates indicate that there are 3.8 million people above 60 years of age in Colombia, which represents 8% of the total population. Most of them are women, living in urban areas and not working. As in many countries in Latin America, the older population in Colombia is increasing rapidly. This trend implies that, whereas today there is one older person for every five children under 15 years of age, in 2025, the ratio will be 1:2. By 2050, the older population is predicted to exceed the child population, when ∼8% of the population is estimated to be aged 80 and over.1–4 These figures suggest that the health and functional status of the Colombian population will be affected by the health of its older citizens, and, as a result, medical care consumption will increase more rapidly than consumption of other goods. The health burden is expected to be further exacerbated by poverty and the increasing prevalence of chronic conditions among younger cohorts who will enter the senior age groups.2

    Mortality and morbidity from injuries among seniors is expected to increase; in particular, traffic crashes and fall injuries will become a larger burden. Worldwide, ∼30% of people above 65 years of age have had a fall in the past year, and close to 20% of these falls ended up requiring hospital care.3 4 It has been documented in developed nations that falls are associated with higher mortality, higher rates of hospitalisation, longer morbidity from hip fractures, and a decline in functioning among seniors—in some cases, with depression.5–7 The development of a fear of falling again can also result in adjustment of physical and social activity.7 Falls are one of the leading causes of death from ‘accidental’ injuries among older people.

    There is a body of literature that relates economic growth to aggregate mortality and morbidity.8–12 The health gains of development have been associated mainly with improvements in nutrition, sanitation, preventive care and housing, which reduce infectious diseases and mortality and morbidity among children.12 Various findings have suggested that economic growth can be positively related to mortality in the short term.8–10 13 Less evidence (most of it in developed nations) is available to document the relationship between economic performance and injury mortality among older people. The work by Moniruzzaman and Andersson14 15 in developed nations indicates that unintentional injuries to people over 75 years increase with economic growth.

    Economic growth could be an attenuating factor for injury mortality among seniors. It may help reduce the hazard of having an injury at home through use of safety devices, help to increase safety on highways and reduce crashes, and provide additional resources for healthcare services that reduce the likelihood of deaths from an injury. However, positive deviation from the trend in growth in gross domestic product (GDP) may also have a negative effect on mortality. GDP expansion may imply a higher number of traffic crashes; furthermore with higher disposable income, the consumption of alcohol and other drugs increases, which may increase the prevalence of injuries. Lastly, economic expansion may create social stress, which may increase the number of intentional injuries among the older population.

    The effect of economic growth on injury among older people is the consequence of a complex web of factors that may work in different directions. If economic expansion alone is not enough to reverse the increasing trends of injury mortality, specific interventions are needed to reduce mortality from preventable injuries among seniors. Clearly, in the absence of economic growth, the rationale to intervene with specific and potentially cost-effective interventions will be even larger.

    In summary, in this paper, we analyse the association between economic growth and injury mortality among seniors in Colombia during the period 1979–2006. It is important to keep in mind that, given the available dataset, we cannot, in practice, measure the causal nature of the association between short-term changes in economic growth and contemporary short-term fluctuations in injury mortality. However, a better understanding of the association between these outcomes may help future research to unravel the causal relationship as well as to provide relevant descriptive information to policy makers. As the impact of economic growth may differ according to type of injury and gender, we conducted the study separately for intentional and unintentional injuries among the older population.

    Data and methods

    Data on injury mortality, economic variables and demographic indicators were obtained from the Colombia National Statistical Agency. For the purpose of this work, we included deaths of people above 60 years of age. In Colombia, the main sources of demographic indicators are Census Projections and The Household Survey. Mortality data were collected from death registration records. Economic variables were obtained from the Colombia National Accounts System. We used fluctuation in real GDP as an economic indicator of economic growth. We also used the population above 60 years of age to control for trend in population growth. For the study period, the Colombian economy grew at an annual average rate of 3.3%, and the population above 60 years of age grew from 1.8 million in 1979 to 3.8 million in 2006.

    Intentional and unintentional injuries were coded according to the International Classification of Diseases (ICD) 9th edition for 1976–1996, and the ICD 10th edition for 1997–2006. From publicly available information, we were able to compile complete injury data and economic information for the period 1979–2006.

    Figure 1 displays the increasing trend in mortality per 100 000 population due to injuries among older people. Before we discuss the observed trend in the data, a word of caution is needed. From the available public information, it is not possible to pinpoint reasons for the peak of deaths after the switch from the 9th edition to the 10th of ICD. However, informal interviews with colleagues in Colombia suggest that the apparent reason for this peak in the trend is merely changes in the recording system. In the regression analysis, we dealt with this unusual break by using a dummy variable approach.

    Figure 1
    Figure 1

    Injury death rate per 100 000 in older people in Colombia, 1979–2006.

    Part of our descriptive approach was to deconstruct basic trends before developing models for further analysis. During the 1976–2006 period, deaths of older people represented ∼51% of total deaths, and injury deaths of older people accounted for 4% of total deaths of older people. Traffic injuries represent the main cause of injury deaths among seniors (31%), followed by homicides (25%) and falls (12%). Table 1 shows details of the trends in leading causes of injury mortality in older people for the period 1979–2006. For all periods, total injury deaths for men were higher than for women; this was also true for deaths from falls, traffic crashes, homicides and suicides. However, for women older than 84 years of age, total deaths due to falls were higher than for men.

    View this table:
    Table 1

    Trends in leading causes of injury deaths in older people from falls, traffic crashes, homicides and suicides by gender and age group, Colombia 1979–2006

    Table 2 summarises the rank order of leading causes of injury mortality in seniors. In both men and women under 84 years of age, traffic crashes were the leading cause of injury deaths; however, for women older than 84 years of age, falls became the main cause of injury deaths.

    View this table:
    Table 2

    Rank order (1–4) of leading causes of injury deaths in older people by gender and age, Colombia 1979–2006

    An initial description of the data suggests a growing trend in the number of injury deaths in seniors for both genders. The leading causes of death differ according to gender, in particular, for seniors above 84 years of age. Taken together, these facts indicate that the association between economic growth and injury mortality differs by gender and cause of injury mortality.

    Empirical strategy

    We implemented four functional forms previously used in work with time-series data. Wooldridge16 provides a detailed explanation of these forms. Given the high expected correlation between the outcomes of interest over years, the autoregressive model seems to be a logical starting point. In this context, it is important to test the stability condition for weak dependence so that ordinary least squares (OLS) estimates and inference can be justified.16 Appropriate statistical tests were performed to decide if the observed time series follow autoregressive models AR(1). Owing to the sample size, however, we were not able to run successful AR(2) models. In addition, the models were run first for all injury deaths, and then separately for each gender. Lastly, we ran the models separately for male and female data for the following types of injury death: falls, traffic crashes, homicides and suicides.

    The four specification forms used in the analysis are as follow:

    1. Static model:Yt=β0+B1GDPt+B2POPt+B3TREND+B4DUMMY_1997+μ

    2. Finite distributed lag model (FDL):Yt=β0+B1GDPt+B2POPt−1+B3POPt+B4TREND+B5DUMMY_1997+μ

    3. First difference modelYt−Yt−1=β0+B1(GDPt−GDPt−1)+B2(POPt−POPt−1)+B3TREND+B4DUMMY_1997+μ

    4. Log modelLogYt=β0+B1logGDPt+B2logPOPt+B3TREND+B4DUMMY_1997+μ

    where Y represents injury deaths, GDP is the real gross domestic product, and POP is the older population (ie, above 60 years of age). We also included in the model a trend variable to control for any trend in the data. In this sense, B1 in model 1 captures the impact on injury deaths of a GDP deviation from its historical trend. We introduced a dummy variable for the year 1997, which accounts for any effect due to the change in coding from the 9th to the 10th edition of ICD. Lastly, μ is the error term.

    We ran models 1–4, both including and without including the trend control variable. We also ran regressions using GDP per capita. However, the findings from these additional specifications (available from AJT upon request) do not differ from the results discussed below.

    Results

    Table 3 shows the results when total injury deaths was used as the dependent variable. In all models (expanded estimations), GDP reduced the total number of deaths. The results are significant at p<0.05 for all specifications. The magnitudes of the GDP effect on injury deaths among the older population are very similar across models. Using the results of the log models implies that, when real GDP increases by 1% above its short-run trend, injury deaths decline by 0.983% even after the observed trend in injuries in older people in the past have been controlled for.

    View this table:
    Table 3

    Ordinary least squares (OLS) results for injury deaths in older people and economic growth, Colombia 1979–2006

    Table 4 shows the results by gender when serial correlation in the error term is controlled for. In this section, we only report the results from the logarithm model. The results based on the other specifications are very similar and are available from the authors on request. Figure 2 displays the real data on injury deaths and the forecast results using the AR(1) log model. The estimations are very close to the real data points. In all cases, we rejected the null hypothesis that sigma is equal to zero, which suggests that serial correlation could be a problem. This specification suggests a higher elasticity, −1.26; however, the result is not significant at p<0.05. Interestingly, GPD growth above historical trend reduces the number of deaths among men while increasing the number of deaths among women.

    View this table:
    Table 4

    Autoregressive model AR(1) for injury deaths in older people and economic growth, Colombia 1979–2006

    Figure 2
    Figure 2

    Injuries and forecast from autoregressive model AR(1), 1979–2006.

    Lastly, Table 5 shows the results when we ran the AR(1) log models separately for each gender using injury deaths as the dependent variable. As before, the results for all specifications are similar and are available from the authors on request. In all cases, we rejected the null hypothesis that sigma is equal to zero. According to the results, an increase in GDP will increase the number of falls; this result is consistent for men and women. However, the result is only significant in the case of women. The results suggest that a 1% increases above the trend in GDP will increase falls by 4.85% among older women. In the case of traffic crashes, an increase in GDP above the trend will result in a reduction in crashes for both men and women. Likewise, GDP growth will reduce the number of homicides and suicides for both men and women.

    View this table:
    Table 5

    Autoregressive models for injury deaths in older people and economic growth, Colombia 1976–2006

    Discussion

    Our findings suggest a moderate association between economic growth and injury deaths among older people, which is line with previous studies in this field.17–20 This relationship is further complicated by the type of injury and gender. Although we find a negative association between GDP growth and injury deaths, the reported elasticity is close to one in most cases. According to the results, economic growth is more effective in reducing traffic deaths, homicides and suicides among older men. This is analogous to the relationship between economic growth and road traffic fatality overall reported by Kopits and Cropper,21 Bishai et al22 and Garg and Hyder.23 Interestingly, economic growth has a positive association with falls among women. A positive association between economic growth and unintentional injuries in developed nations has been reported12 for people 75 years or older.

    Using longitudinal data, Moniruzzaman & Andersson14 reported that at a low level of GDP, injury mortality increases, then declines steadily. According to these authors, road traffic injuries and falls decline at early stages of the developmental process. It may be that our findings on falls using a time-series analysis reflect the initial portion of the U-shaped trend in economic growth and injury-related mortality in middle-income countries.

    Gender differences in injury-related behaviour and mortality have been reported previously.23 Given the short duration of the time series data, we could not test if the observed gender gaps in injuries are part of the larger process of convergence of the genders as reported in developed nations. However, the information suggests room to introduce a differentiated set of health policies by gender to reduce injury-related deaths among seniors.

    The rapid ageing process in Colombia suggests that mortality and morbidity among the older population from injury can be expected to increase; in particular, traffic crashes and fall injuries will be a larger burden in the health of seniors. The results suggest that the impact of economic growth on injury deaths among older people is not homogeneous by gender or type of injury. Policy makers may need to combine macroeconomic policy with specific health interventions to reduce the burden of injury in the older population.24–26 This is a challenge that a developing country, such as Colombia, must face as their epidemiological transition drives a shift from transmissible/neonatal priorities towards injury and other non-communicable death causes. Interestingly, the long-term development plan of the country's health sector (Vision Colombia 201927) does not consider social investment on preventive and rehabilitative programmes to reduce the burden of injuries among seniors.

    The range of interventions that have been shown to be effective in preventing falls in both community dwelling and institutionalised older people include: muscle strengthening and balance retraining; tai chi group exercise; home hazard assessment and modification.28 29 Reduction of psychotropic drugs and supplementation with vitamin D may also reduce falls.30 31 Hip protectors, while effective, have poor compliance rates.32 Effective strategies for middle-income countries such as Colombia may include these or new approaches that require innovative research.

    Conclusion

    Injury is an important public health problem with enormous impact on medical care consumption, disability and mortality among seniors. Reducing preventable injuries among seniors would improve quality of life and reduce the number of injury-related deaths. Economic growth may not be enough to reduce the trend in injury mortality among seniors. The results suggest a non-homogeneous moderate effect of growth on death of older people. This paper contributes to the literature in the injury field in two fundamental ways. Firstly, it investigates the association between macroeconomic growth and injury deaths among the older population. Secondly, it analyses this association in the context of a middle-income country with a rapid ageing process.

    These results should be interpreted with caution. Colombia is a middle-income economy with an unstable macroeconomic environment because of security and political tension. The years included in the regression were dominated by some political and institutional progress compared with previous decades. This provided sustainability to the observed economic growth. In fact, recent fiscal adjustment and security improvements have raised confidence in Colombia's prospects. However, the observed association between injury and growth may be specific to the local situation and difficult to extrapolate to other countries.

    Further research is needed to understand the underlying factors behind a possible causal relationship between growth and injury, and to explain the gender differences. One can expect the trend to follow the convergence hypothesis.33 However, a better understanding of the process is necessary to develop effective injury policies.

    What is already known on this subject

    • The existing literature indicates that economic growth reduces injury mortality among children. This result is consistent in developed and middle-income countries.

    • Less is known about the association between growth and injury mortality among seniors. Studies in developed countries suggest a weak relationship.

    What this study adds

    • An analysis of injury among older people in a middle-income country (Colombia).

    • Evaluation of the association between economic growth and injury by type of unintentional injury and by gender.

    • More appropriate time-series modelling than in existing studies on developed societies.

    References

    1. ↵
      1. Todd J,
      2. Rossi M,
      3. Triunfo P
      . The Health Status of Elderly Populations in Latin American Countries. Cuadernos de Economia 2007;26:141–7.
      OpenUrl
    2. ↵
      World Health Organization. WHO Global Report on Falls Prevention in Older Age. Report. Geneva, Switzerland: World Health Organization, 2007.
    3. ↵
      World Health Organization. The injury chart book: a graphical overview of the global burden of injuries. Report. Geneva, Switzerland: World Health Organization, 2002.
    4. ↵
      Pan American Health Organization. The State of Aging and Health in Latin America and the Caribbean. Washington: PAHO, 2008.
    5. ↵
      1. Harlan LC,
      2. Harlan WR,
      3. Parsons PE
      . The economic impact of injuries: a major source of medical costs. Am J Public Health 1990;80:453–9.
      OpenUrlPubMedWeb of Science
    6. ↵
      1. Kannus P,
      2. Parkkari J,
      3. Niemi S,
      4. et al
      . Fall-induced deaths among elderly people. Am J Public Health 2005;95:422–4.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Li W,
      2. Keegan TH,
      3. Sternfeld B,
      4. et al
      . Outdoor falls among middle-aged and older adults: a neglected public health problem. Am J Public Health 2006;96:1192–200.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Birchenall JA
      . Economic development and the escape from high mortality. World Development 2007;35:543–68.
      OpenUrlCrossRefWeb of Science
    9. ↵
      1. Brenner MH
      . Commentary: economic growth is the basis of mortality rate decline in the 20th century–experience of the United States 1901–2000. Int J Epidemiol 2005;34:1214–21.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Kunitz SJ
      . Commentary: Samuel Preston's 'The changing relation between mortality and level of economic development'. Int J Epidemiol 2007;36:491–2; discussion 502–3.
      OpenUrlFREE Full Text
    11. ↵
      1. McKeown T
      . The modern rise of population. New York: Academic Press, 1976.
    12. ↵
      1. Preston SH
      . The changing relation between mortality and level of economic development. Popul Stud (Camb) 1975;29:231–48.
      OpenUrlCrossRefPubMed
    13. ↵
      1. Granados JAT
      . Increasing mortality during the expansions of the US economy, 1900–1996. 2005:1194–202.
    14. ↵
      1. Moniruzzaman S,
      2. Andersson R
      . Relationship between economic development and risk of injuries in older adults and the elderly. A global analysis of unintentional injury mortality in an epidemiologic transition perspective. Eur J Public Health 2005;15:454–8.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Moniruzzaman S,
      2. Andersson R
      . Economic development as a determinant of injury mortality - a longitudinal approach. Soc Sci Med 2008;66:1699–708.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Wooldridge JM
      . Introductory Econometrics: A Modern Approach. 2nd edn. Thomson South-Western.
    17. ↵
      1. U S Gannon B,
      2. O'Shea E,
      3. Hudson E
      . Economic consequences of falls and fractures among older people. Ir Med J 2008;101:170–3.
      OpenUrlPubMed
    18. ↵
      1. Ling C,
      2. Henderson S,
      3. Henderson R,
      4. et al
      . Cost benefit considerations of preventing elderly falls through environmental modifications to homes in Hana, Maui. Hawaii Med J 2008;67:65–8.
      OpenUrlPubMed
    19. ↵
      1. Royan SJ,
      2. Hamid AL,
      3. Kovilpillai FJ,
      4. et al
      . A prospective study on elderly patients with facial fractures in a developing country. Gerodontology 2008;25:124–8.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Sari AB,
      2. Cracknell A,
      3. Sheldon TA
      . Incidence, preventability and consequences of adverse events in older people: results of a retrospective case-note review. Age Ageing 2008;37:265–9.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Kopits E,
      2. Cropper M
      . Traffic fatalities and economic growth. Accid Anal Prev 2005;37:169–78.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Bishai D,
      2. Quresh A,
      3. James P,
      4. et al
      . National road casualties and economic development. Health Econ 2006;15:65–81.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Garg N,
      2. Hyder AA
      . Exploring the relationship between development and road traffic injuries: a case study from India. Eur J Public Health 2006;16:487–91.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Hendriks MR,
      2. Evers SM,
      3. Bleijlevens MH,
      4. et al
      . Cost-effectiveness of a multidisciplinary fall prevention program in community-dwelling elderly people: a randomized controlled trial (ISRCTN 64716113). Int J Technol Assess Health Care 2008;24:193–202.
      OpenUrlPubMedWeb of Science
    25. ↵
      1. Shigematsu R,
      2. Okura T,
      3. Nakagaichi M,
      4. et al
      . Square-stepping exercise and fall risk factors in older adults: a single-blind, randomized controlled trial. J Gerontol 2008;63:76–82.
      OpenUrlWeb of Science
    26. ↵
      1. Zecevic AA,
      2. Salmoni AW,
      3. Lewko JH,
      4. et al
      . Seniors Falls Investigative Methodology (SFIM): a systems approach to the study of falls in seniors. Canadian journal on aging = La revue canadienne du vieillissement. 2007;26:281–90.
    27. ↵
      National Planning Department. Vision Colombia II. Centenario: DNP, 2008.
    28. ↵
      1. Chang JT,
      2. Morton SC,
      3. Rubenstein LZ,
      4. et al
      . Interventions for the prevention of falls in older adults: systematic review and meta-analysis of randomised clinical trials. BMJ (Clinical research 2004;328:680. ed.
      OpenUrlCrossRef
    29. ↵
      1. Gillespie LD,
      2. Gillespie WJ,
      3. Robertson MC,
      4. et al
      . Interventions for preventing falls in elderly people. Cochrane database of syst rev (Online) 2003(4):CD000340.
    30. ↵
      1. Bischoff-Ferrari HA,
      2. Dawson-Hughes B,
      3. Willett WC,
      4. et al
      . Effect of Vitamin D on falls: a meta-analysis. JAMA 2004;291:1999–2006.
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. Campbell AJ,
      2. Robertson MC,
      3. Gardner MM,
      4. et al
      . Psychotropic medication withdrawal and a home-based exercise program to prevent falls: a randomized, controlled trial. Journal of the American Geriatrics Society 1999;47:850–3.
      OpenUrlPubMedWeb of Science
    32. ↵
      1. Parker MJ,
      2. Gillespie LD,
      3. Gillespie WJ
      . Hip protectors for preventing hip fractures in the elderly. Cochrane database of syst rev (Online) 2003;(3):CD001255.
    33. ↵
      1. Waldron I,
      2. McCloskey C,
      3. Earle I
      . Trends in gender differences in accidents mortality: relationships to changing gender roles and other societal trends. Demographic Research 2005;13:415–54.
      OpenUrlCrossRefWeb of Science

    Footnotes

    • Competing interests None.

    • Provenance and peer review Not commissioned; externally peer reviewed.