Mortality rate and years of life lost from unintentional injury and suicide in South India

Authors


Corresponding author Anuradha Bose, Department of Community Health, Christian Medical College, Vellore, India 632002. Tel.: 91 (0) 416 2284207; Fax: 91 (0) 416 2262268; E-mail: abose@cmcvellore.ac.in

Summary

We calculated mortality rates and years of life lost because of unintentional injuries and suicides using community based information obtained prospectively over a 7-year period, from 1998 to 2004, among a rural and peri-urban population of 108 000 in South India. Per 100 000 population the total mortality rate for unintentional injuries and suicides combined was 137.1, with 54.9 for unintentional injuries and 82.2 for suicides respectively. Hanging and self-poisoning with pesticides were the preferred means of suicide. Unintentional injuries and suicides resulted in 26.9% of total life years lost over the study period while 18.9% of all deaths in the population were attributable to unintentional injuries and suicides in the same period. The high burden is particularly notable in the 15–29 age group, where up to 70% of years of life lost are due to injury. The burden of injuries reported in this study is significantly higher than the figures reflected in available reports for India and is likely due to the under reporting in routine mortality statistics, particularly of suicides.

Abstract

Le taux de mortalité et le nombre d'années de vie perdues dus aux blessures involontaires et aux suicides ont été calculés à partir d'une base d'information communautaire obtenue prospectivement pendant une période de sept ans, de 1998 à 2004, parmi une population rurale et péri-urbaine de 108 000 habitants en Inde du Sud. Pour 100 000 habitants, le taux de mortalité total combinant blessures involontaires et suicides était de 137.1 avec respectivement 54.9 pour les blessures involontaires et 82.2 pour les suicides. Les pendaisons et les auto-empoisonnement par pesticides étaient les moyens de suicides prédominants. Le nombre d'années de vie perdues sur la période étudiée par blessures involontaires et suicides était de 26.9% alors que 18.9% de tous les décès dans la population était attribuables à des blessures involontaires et à des suicides sur la même période. Ce fléau était particulièrement notable dans le groupe d’âge 15–29 ans, où plus de 70% d'années de vie perdues étaient dues à des blessures. Le poids des blessures rapportées dans cette étude est significativement plus élevé que les chiffres des rapports disponibles pour l'Inde. Ceci est probablement dû aux sous déclarations des décès et particulièrement des suicides.

Abstract

Los índices de mortalidad y los años de vida perdidos debido a lesiones no intencionales y suicidios fueron calculados utilizando información de base comunitaria obtenida prospectivamente en un período de siete años, desde 1998 a 2004, entre una población rural y suburbana de 108.000 personas en el sur de la India. Para una población de 100.000 personas, la tasa total de mortalidad por lesiones no intencionales y suicidios combinada fue de 137.1, con 54.9 para lesiones no intencionales y 82.2 para suicidios respectivamente. Ahorcamiento y auto envenenamiento con pesticidas fueron las formas preferidas de suicidio. Lesiones no intencionales y suicidios dieron como resultado un total de un 26.9% del total de años de vida perdidos durante el período de estudio, mientras que un 18.9% de todas las muertes en la población fueron atribuidas a lesiones no intencionales y suicidios en el mismo período. El peso mayor es particularmente notable en el grupo entre 15 y 29 años, donde hasta el 70% de los años de vida perdidos se deben a lesiones. El peso de las lesiones reportadas en este estudio es significativamente mayor de lo que reflejan las cifras disponibles de India, y es probable que sea debido a la falta de denuncias en las estadísticas rutinarias de mortalidad, particularmente los suicidios.

Introduction

In India, it is estimated that over 1.2 million are injured seriously, about 300 000 are disabled permanently and more than 80 000 people die in road traffic injuries annually (Mohan 2004). However, the burden caused by injuries is not well established in low- and middle-income countries and the surveillance systems are poor (Krug 2004; Mock et al. 2004). Such data are vital for planning prevention efforts and to optimize the distribution of available resources. Much of the data that are available is from hospital statistics and police records, which could lead to an under-representation of the problem or a misrepresentation of the causes of injury especially in rural areas. The data presented here, we believe, are representative of the actual mortality caused by unintentional injuries and suicides among the rural and peri-urban populations of South India. Our results add to earlier reports from the same community by providing information from the past 7 years for all age groups and for both unintentional injuries and suicides (Joseph et al. 2003; Aaron et al. 2004; Abraham et al. 2005), and by adding years of life lost because of injury.

Methods

The findings presented here are the results of an ongoing surveillance system established by the Christian Medical College, Vellore, India, covering a population of approximately 108 000. A computerized health information system has been in operation since 1986 which prospectively records all information on births and deaths within the community. The mortality information is based on the use of the verbal autopsy method (Aaron et al. 2004).

The surveillance system is tiered, with community health workers living in the villages having access to detailed and reliable information about the circumstances of death. When a death occurs in the village, its details and circumstances are assessed by the village health workers, the village nurse and the doctor assigned to that particular village and a consensus is reached on the nature of death. In the event that the nature or cause of death is unclear, a doctor visits the village and speaks to members of the family or to other informed members of the community to clarify details. Annually a 10% validation of all data obtained is done by independent observers from the Department of Community Health.

We analysed the mortality for the 7 years from 1998 to 2004. Unintentional and intentional injuries were categorized by age and sex for all ages from the post-neonatal period onwards. Intentional injuries reported here include self-inflicted suicides. Unintentional injuries include road traffic injuries, drowning, falls, fires and unintentional poisonings. Information relating to 57 homicides and 30 cases of death because of unintentional occupational blast injuries were omitted from the analysis as detailed information could not be obtained for these cases through the established procedure of verbal autopsy. To further validate information for these cases access to sensitive police investigations and legal hearings would have to be pursued and fell outside the scope and ethical clearance of the study. However, for all the excluded cases the verbal autopsy clearly indicated that they were not associated with self-harm. Years of life lost were calculated from standard WHO life-tables using a discount rate of 3% and standard age weights, for all ages. Place of death was categorized as ‘death at home or at the site of the incident’, ‘death while on the way to hospital’ and thirdly as ‘death at the hospital’.

Results

Table 1 shows details of death from unintentional injury and suicide by gender and by age category. Per 100 000 population the total mortality rate for unintentional injuries and suicides combined was 137.1, with 54.9 for unintentional injuries and 82.2 for suicides respectively.

Table 1.   Prevalence of suicide and injury in Kaniyambadi block (1998–2004)
 0–4 years5–14 years15–29 years30–44 years45–59 years60–69 years70–79 years≥80 years
SuicideInjurySuicideInjurySuicideInjurySuicideInjurySuicideInjurySuicideInjurySuicideInjurySuicideInjury
  1. No cases of suicide was registered for individuals below 10 years of age.

  2. Injury here is defined as all categories of unintentional injuries.

Male
 No of deaths017427965792528644404432311924
 Person years28 24465 921103 33592 48352 54520 66110 8213585
 Rate per 100 0000.060.26.141.092.955.299.556.2163.783.7193.6213.0295.7286.5530.0669.5
 % of total deaths0.06.66.644.342.925.424.814.015.57.97.48.24.84.73.13.9
 All cause of death25661224371554538663616
Female
 No. of deaths0155912312465301629142735924
 Person years26 60565 230123 25088 83852 54524 44313 3274235
 Rate per 100 0000.056.47.713.899.89.751.85.657.130.5118.657.3202.6262.6212.5566.7
 % of total deaths0.05.815.628.158.05.731.53.411.76.29.74.75.46.91.43.7
 All cause of death25732212146257300504650

Suicide constituted 11.3% of all deaths across all age groups in the study population. Of the 638 suicide cases registered over the 7 years of monitoring, hanging was the preferred method with 305 cases followed by the use of poison (258), burning (46), drowning (25) and various other means making up only four cases. No gender difference was found in preference for hanging but more men preferred poison and more females preferred drowning and self-immolation as a means of suicide. Relative to other means of suicide, hanging became more frequent over the 7 years of monitoring (χ2 for trend; P < 0.05), whereas the relative frequency of poisoning (P > 0.05) and self-burning (P > 0.05) remained unchanged over the period. Drowning became less common relative to other means (P < 0.05). The majority of suicide poisoning cases followed the consumption of pesticides, including both Class I and II pesticides. Approximately 70% of the deaths because of suicide occurred at home or at the site of self-harm. The methods reported here reflect only those incidents of self-harm which resulted in death, as morbidity data are not collected routinely. However, it is important to emphasize that the methods of suicide reported here may not be a reflection of the actual prevalence of methods for attempted suicide, but more likely reflect the lethality of the methods chosen.

Unintentional injuries contributed to 7.6% of all deaths across all age groups in the study population. In all ages, more men than women died as a result of unintentional injury. Amongst the age group below 15 years and for the age group above 70, unintentional injury killed more than suicide when taking men and women combined. In all other age groups, suicide ranked higher. Of the 426 cases of death from unintentional injury the two most important were road traffic injuries with 137 deaths and unintentional drowning with 87 deaths. More men than women died from road traffic injuries.

A total of 6134 life years were lost each year in this population of 108 000 because of unintentional injuries and suicides. This figure was determined by calculating life years lost in each age group and dividing that number by seven to arrive at cumulative life years lost for the population each year. This contributes to 26.9% of total life years lost over the study period, while 18.9% of all deaths were attributable to mortality from unintentional injuries and suicides. The proportional distribution of years of life lost because of unintentional injury and suicides is shown for men (M) and women (F) in Figure 1. The graph highlights that injuries tend to affect younger people, in the most productive period of their lives.

Figure 1.

 Proportion of years of life lost because of unintentional injuries and suicide by age and sex categories.

Discussion

Tamil Nadu Government statistics report that the total deaths because of injuries alone in the state were 10 717, or 2.9% of all reported deaths (Government of India 2006). The findings of this study indicate a higher burden associated with injuries and record a very high rate for suicides. The high mortality from injuries reported here is also a reflection of the differences in quality of statistics obtained from detailed community based reporting and the data that is captured by the Government reporting system, especially with respect to information on suicides where there may be hesitation in disclosing information to the authorities. The size and diversity of India precludes us from assuming that these rates would be found in other parts of India. There is, however, nothing different sociologically or ecologically in this area from the rest of Tamil Nadu. We believe that if the surveillance system were duplicated, higher rates would be reported from other areas in South India. The mortality and injury rates call for immediate action from society at large and the medical establishment and public health authorities in particular, to reduce the burden caused especially by suicides by hanging and poisoning and to reduce road traffic injuries and unintentional drowning.

Acknowledgements

We acknowledge the efforts of the entire team of the Department of Community Health in the collection and verification of this data and for ensuring its accuracy.

Ancillary