Potential Gain of Life Expectancies for Elimination of Leading Causes of Death in Bangladesh

Md. Aminul Hoque*
Department of Statistics, University of Rajshahi

Correspondence to:
Md. Aminul Hoque
Ph.D., Associate Professor,
Department of Statistics,
University of Rajshahi,
Rajshahi-6205, Bangladesh
Mobile: +88-1914254017
Fax: +88-0721-750064-42241 (Ext).
Email: mdaminulh@gmail.com

INTRODUCTION

At present the world is anxious to increase life expectancy. Already the developed countries have increased their life expectancy by controlling and eliminating the vulnerable diseases. But the underdeveloped and developing countries like Bangladesh still have low life expectancy. A hundred years ago, the average life expectancy in most parts of the world was 25 to 30 years. According to the WHO (1997-1999) and World Health Statistics Annual (see also the World Health Report 1998), life expectancy averages 64 years in the developing nations and is approaching 80 years in some industrial nations. Today in most developed countries it has stretched to over 70 years (Kabir 1987). The expectation of life at birth is 62 years for national cases of Bangladesh (Gordon 2003) and the expectation of life at birth is 61.7 years from national cases of Bangladesh (SASNET 2006). According to the "Health Department of United Nation" the cause of death is social and (not only) biological (Gordon 1956). the majority of deaths of developing countries like Bangladesh are due to different types of disease.

A large number of mortality and morbidity cases are due to communicable diseases. Communicable diseases were the major health hazards in the 1980s. Some prime communicable diseases are diarrhoea, dysentery, tuberculosis, meningitis, hepatitis, chicken pox and rabies. Again cardiovascular disease (hypertension ischaemic heart disease and stroke) is very common cause of . AIDS is responsible for mortality in western countries but ir is now also at an alarming rate in developing and underdeveloped countries. Respiratory disease is also a very big problem for the elderly and children too.

A joint statement (WHO and UNDP, 1986) emphasizes that acute respiratory infection, diarrhoeal diseases and malnutrition are the principal causes of illness and death of children in developing countries. Disease in the late 1980s was most prevalent in rural areas; treatment was more readily available in the cities. A mid-1980s survey indicated that deaths due to diarrhoeal diseases, malnutrition, and pneumonia accounted for 16.3 percent, 13.1 percent, and 10.8 percent of all deaths, respectively. The percentages for other diseases were as follows: prematurebirth and birth injury (8.6 percent), cardiovascular accidents (4.5 percent), tetanus (4.4 percent), pulmonary tuberculosis (3.3 percent), measles (2.7 percent), and other causes (36.3 percent) (Hoque and Islam, 2002).

Most of the world is on target to reduce the impact of tuberculosis (TB) but efforts have yet to bear fruit in Africa where the disease goes hand in hand with AIDS (UN, 2006). In its annual report on the global impact of TB, the WHO said that the disease claimed 1.7 million lives in 2004 and that there were almost 9 million new infections. Tuberculosis bacteria infect more than three hundred thousand people and around 70,000 people die every year in the country. In another place one person is infected every two minutes and is dying every ten minutes of tuberculosis in Bangladesh. The WHO said that two of the worst affected countries, Philippines and Vietnam had already met the champion target. Bangladesh stands at fifth position among 22 high TB prevalence countries in the world while it ranks fourth among five leading TB-prone countries in South East Asia, according to World HealthOrganization. Dr Khurshid Alam Hyder, WHO consultant, presented a paper on the global and Bangladesh TB situation. He said TB bacteria have infected one-third of the world population. TB infects eight million people every year and claims two million lives yearly throughout the world.

Diarrhea is responsible for high morbidity and mortality in children and adults in developing countries (George et al. 1997). Diarrhoeal diseases still hold the second rank in the causes of global mortality and morbidity amongst infants in the developing world. Children aged less than 1 year, are the most vulnerable group (S.Dutta et al.1997). Globally an estimated four million children and adults die anally of infectious diarrhea (Barua D.1981). A high-risk diarrhea occurs when babies are on complementary feeding (Synder et al 1982). Sample Vital Registration System (SVRS) 1999-2002, BBS estimates 7.02 percent people of Bangladesh died of total deaths from cholera, diarrhea and dysentery which includes the diarrhoeal group in 2002. From Health and Demographic Surveillance System (HDSS) - Matlab data published by ICDDR, B we have to estimate 5.16% and 4.85% people died of total deaths from diarrhoeal diseases in 2000 and 2003 respectively.

Acute Respiratory Infection (ARI) is the most common cause of illness in children and a major cause of death in the world. Among children under 5 years of age 3 to 5 million deaths annually have been attributed to ARI of which 75% are from pneumonia. The World Health Organization (WHO) estimates that approximately three million children under five years died from ARI in 1993. ARI is also a significant cause of death at other ages particularly in the very old. Sample Vital Registration System (SVRS) 1999-2002, BBS estimates 13.92 percent people of Bangladesh died of total deaths from asthma and respiratory diseases in 2002. From Health and Demographic Surveillance System (HDSS) - Matlab data published by ICDDR, B we have to estimate 12.28% and 14.42% people died of total deaths from respiratory disease in 2000 and 2003 respectively.

Injuries of different natures and degrees are common health problems in everyday life. Especially tetanus, a fatal disease, is quite common in our country. Tetanus is the third cause of high infant mortality in Bangladesh (Khan, 1986). Sample Vital Registration System (SVRS) 1999-2002, BBS estimates 3.71 percent people of Bangladesh died of total deaths from tetanus in 2002.

Another disease viz. cardiovascular is the greatest killer in western countries. With the control of infectious and other diseases in our country, cardiovascular diseases are going to be the number one problem in the near future (Malik, 1979). Sample Vital Registration System (SVRS) 1999-2002, BBS estimates 10.51 percent people of Bangladesh died of total deaths from blood pressure, heart disease, and stroke which is included in cardiovascular disease, in 2002. From Health and Demographic Surveillance System (HDSS) - Matlab data published by ICDDR, B we have to estimate 12.15% and 19.46% people died of total deaths from Cardiovascular diseasein 2000 and 2003 respectively.

For the diseases called causes of death mentioned above, like other undeveloped countries the expectation of life is still low in rural populations. If we are able to eliminate only one killer disease such as diarrhea, infectious or respiratory disease etc. from our country, then our life duration will be increased sharply. Dr. Farr's work was published in the Supplement to the Thirty-Fifth Annual Report of the Registrar General, 1875 and entitled.

Effect of the Extinction of any single disease on the duration of life'. Farr attempted the problem of eliminating a particular disease from the life table population, taking among other diseases, as example cancer and phthisis. Karn (1931) in his paper entitled 'An inquiry into various death rates and the comparative influence of certain Disease on the Duration of Life' constructed a life table for a population from which cancer and tuberculosis were supposed to be eliminated as causes of death in order to estimate the effect of these diseases in shortening the duration of life. We have come to know from Karn's (1993) paper that before the Farr's work the French Mathematician Daniel Bernoulli and D'Alembert found the increase of mean life time which would result from the extinction of small-pox. They discussed the expected consequences on mortality model if small pox were to be eliminated as a cause of death.

It is well known that the development indicator is the disease free life. Elimination of a cause of death has real natural meanings. Removal of the sources of infection or prevention of fetal development of the disease is conducted by vaccination. At present diarrhoeal and infectious diseases associated with aging have now become the major causes of death in Bangladesh. It is not that difficult to eliminate diarrhoeal and most of the infectious diseases. Constructing multiple decrement life tables with associated single decrement life tables provides scope to explore probability distribution or pattern of deaths by age due to different causes acting simultaneously in the population. Hoque (1993) carried out a such study using multiple decrement technique. Abedin and Islam (1993) have analyzed age and cause specific mortality data by constructing multiple and single decrement life tables in order to see the probability distribution by age, for water related diseases which seem to be one of the major causes of death of Matlab population, Bangladesh. Hoque and Abedin (1996) estimated the cause specific life expectancy of the elderly population of Matlab. Very recently Hoque (2008) and Hoque and Islam (2002) estimated the effect of elimination of a specific disease on life expectatancy in Bangladesh and construct a single decrement life table.

In this paper, an attempt has been made to study the effect of elimination of diarrhoeal and infectious diseases separately and combinedly, causes of death on probabilities and life expectancies of the rural (Matlab) population of Bangladesh for the years 2000 and 2003.

MATERIALS AND METHODS

For the purpose of this study, age and cause specific mortality data are required. The Health and Demographic Surveillance System (HDSS) of Matlab data on vital events by age and cause specific death, published by ICDDR, B, appears to be the source available for this purpose. Since 1966 the Matlab area has had a continuous mortality registration system. The present study uses 2000 and 2003 mortality data by age and cause of death in the scientific report no. 89 and 92 respectively.

The age specific mortality rate (_nM_x) is computed by dividing the observed death (nDx) by the mid year population (_nP_x) i.e. _nM_x=_nD_x/_nP_x and converted into probability (conditional) death in (x, x + n) denoted by nqx given alive age x on the assumption that deaths are uniformly distributed over the age interval (x, x + n).
The conversion formula is _nqx=2_n. _nM_x /(2+n. _nM_x)

The above formula is used to compute conditional probabilities of death from all causes excluding diarrhoeal diseases, excluding infectious diseases and excluding diarrhoeal & infectious diseases simultaneously. In order to compute the peculiarities of death due to diarrhoeal diseases as well as infectious diseases and diarrhoeal and infectious diseases combinedly in presence of all other causes of death, a single decrement life table isconstructed. For causes excluding diarrhoeal diseases, excluding infectious diseases and excluding diarrhoeal & infectious diseases combined, are labeled as (-1), (-2) and (-3) respectively. The conditional probability of dying in age interval (x, x + n) for all causes, excluding diarrhoeal diseases, excluding infectious diseases and excluding diarrhoeal & infectious diseases combinedly are denoted by nqx, _nq(-1_x), _nq(-2_x), and _nq(-3_x) respectively be computed the conditional probabilities given in the multiple decrement life tables.

Once _nq_x and _nq_(-ix), (i= 1, 2 & 3) are obtained the survival functions lx are taken of
l0 =100000 and using the relationship
l_x+n= l_x´(1- _nq_x) and
l_(-i.x+n) = l_(-ix)´(1- _nq_(-ix)); i = 1, 2 & 3.
Person - years lived between ages x to x + n is denoted by nLx which can be defined by
_nL_x= 0.3l_x + 0.7l_x+n; for x = 0 and n = 1,
_nL_x= 0.4l_x + 0.6l_x+n; for x = 2 and n = 4,
_nL_x= n/2´(l_x + l_x+n); for x = 5, 10, 15, …, 80 and n = 5 and
_nL_x = l_x/_nM_x; for x =85+ and n is opened.

Total number of person - years lived above age x is denoted by T_x which can be defined by
T_x= _xL_y; where x is the lower limit of age group.

The expectation of life denoted by e_x⁰ , which can be defined by
e_x⁰ = T_x/l_x; where x is the lower limit of age interval.

The life expectancy in age interval (x, x + n) for all causes, excluding diarrhoeal diseases, excluding infectious diseases and excluding diarrhoeal & infectious diseases combined are denoted by e_x⁰, e⁰_(-1x), e⁰_(-2x), and e⁰_(-3x), respectively given in the multiple decrement life tables.

Again the additional expectation of life denoted by a(ix) is computed from the difference between all causes and excluding a specific cause that is
a_(ix) = e⁰0 - e⁰_(-ix); i = 1, 2 & 3.
Having obtained the life expectancies due to these causes, they are computed in the usual manner (Johnson, 1980).

RESULTS AND DISCUSSION

We are interested to examine the death pattern and expectation of life due to all causes except diarrhoeal diseases, all causes except infectious diseases and also due to all causes except diarrhoeal and infectious diseases combined. For the purpose we have constructed single decrement life tables eliminating a specific disease one after another from all acting diseases.

Table1: Probabilities of dying for excluding a specific cause for the year 2000 and 2003 of Matlab population:

Table 1 gives the probabilities of dying for all causes with excluding a specific disease separately and combinedly of Matlab population for the years 2000 and 2003. From this table, we have found that the risks of death by all causes are relatively higher than by all causes except diarrhoeal disease as well as except infectious disease. The same results followed by eliminating diarrhoeal and infectious diseases combinedly over all ages for both the years 2000 and 2003 of Matlab population.

It has also been found that the risks of death due to different causes markedly declines for the year 2003 than that of 2000 for all age groups. Naturally the infant probability of dying is greater than that of older age probabilities of dying for both the calendar years 2000 and 2003. It is found from this table that the risks of death of middle year age are relatively low compared to that of early and elderly age due to all causes for both the study years exhibiting the typical mortality pattern. In estimating the expectation of life and the additional expectation of life resulting from elimination of a specific cause of death by normal method, are set out in Table 2.

Table 2 - Expectation of life due to all causes and excluding a specific cause for the year 2000 2003 of Matlab population:

From Table 2 and Figures 1-3, we have found that for the elimination of diarrhoeal diseases the ordinary expectation of life at birth is 65.29 years which is increased 0.84 years than for all causes in 2000, by the elimination of infectious diseases the expectation of life is 65.30 years which is increased by 0.85 years than for all causes and by the elimination of diarrhoeal and infectious diseases combinedly the expectation of life at birth is 66.18 which is increased by 1.73 years than for all causes in 2000.

Figure 1: Additional Expectation of Life at birth of Matlab Population in 2000

Figure 2: Additional Expectation of Life at birth of Matlab Population in 2003

These differences became larger in the former ages than the latter ages. It also indicates that the differences in expectation of life ranging from 0.84 years at birth to 0.74 years in the case of diarrhoeal diseases, from 0.85 years at birth to 0.05 years in case of infectious diseases and 1.73 years at birth to 0.81 years in case of diarrhoeal and infectious diseases combinedly in 2000.

Figure 3: Comparative Pattern of the Additional Expectation of life at Birth between 2000 and 2003 of Matlab Population

Again we have found that for the elimination of diarrhoeal diseases the ordinary expectation of life at birth is 66.34 years which is increased by 0.67 years than for all causes in 2003, by the elimination of infectious diseases, the expectation of life is 66.37 years which is increased 0.70 years than for all causes and by the elimination of diarrhoeal and infectious diseases combinedly the expectation of life at birth is 67.10 which is increased 1.43 years than for all causes in 2003. It has been noticed that the expectation of life at birth increased 2.86 years in the Matlab population if we eliminate the diarrhoeal & infectious diseases combined (Hoque and Islam 2002). Also it has been noticed that the expectation of life at birth increased 5.50 years from national cases if we eliminate the diarrhoeal diseases only (BBS, 1996).

It also indicates that the differences in expectation of life ranging from 0.67 years at birth to 0.57 years in case of diarrhoeal diseases, from 0.70 years at birth to 0.04 years in case of infectious diseases and 1.43 years at birth to 0.63 years in case of diarrhoeal and infectious diseases combinedl in 2003.

Table 2 and Figure 3 also show that the expectation of life at birth is increased by 1.22 years in the case of all causes of death, 1.05 years in case of diarrhoeal diseases elimination, 1.07 years in the case of infectious diseases elimination and 0.92 years in the case of diarrhoeal & infectious diseases elimination from the calendar year 2000 to 2003 of the Matlab population.

From Figures 1, 2 and 3 we have found that the expectation of life by elimination of both diarrhoeal and infectious diseases is always greater than the expectation of life for all causes and also by elimination of diarrhoeal and infectious diseases separately the expectation of life is greater than the expectation of life for all causes in 2000 and 2003.

Again from Figure 3, obviously we say that the expectation of life at birth by elimination of diarrhoeal and infectious diseases separately and combinedly in 2003 is greater than from 2000.

Again from Figures 1-3 obviously we show that the additional expectation of life by elimination of both diarrhoeal and infectious diseases is greater than the additional expectation of life by elimination of diarrhoeal and infectious diseases separately in 2000 and 2003. And we have to show that the additional expectation of life at former ages is greater than the latter ages by elimination of diarrhoeal and infectious diseases separately and combinedly in 2000 and 2003. We have also found that the additional expectation of life in 2000 is greater than the additional expectation of life in 2003 in the case of elimination of diarrhoeal and infectious diseases separately and combined.

To compare the cause specific and year specific life expectancy at birth and % of death we estimate Table 3.

Table 3 - Expectation of life at birth and percent of death due to specific causes of Matlab population for 2000 and 2003:

Figure 4: Comparative Pattern of the Expectation of Life at Birth for the Years 2000 and 2003 of Matlab Population

Figure 5: Comparative pattern of the Percentage of Deaths for the Years 2000 and 2003 of Matlab Population

From Table 3, we have to show that the expectation of life at birth is increased in 2003 than 2000 for all causes. It has been found that 5.16% of the total deaths occurred due to diarrhoeal diseases for the year 2000 whereas, 4.85% for the year 2003 are due to the same diseases. It is obviously noticed that the death due to diarrhoeal diseases sharply declined since last decade in Matlab population. For this reason the expectation of life at birth excluding diarrhoeal diseases increased by 1.05 years from 2000 to 2003. Similar trends have been found for excluding infectious diseases and diarrhoeal & infectious diseases combined.

From Figure 4, we obviously see that the expectation of life at birth for different causes is increased in 2003 than 2000, therefore we have to say that the expectation of life is going to be increased in the Matlab population. Again from Figure 5, we have found that the % of deaths due to different causes is decreased in 2003 than 2000. In this case the expectation of life at birth is increased from 2000 to 2003.

CONCLUSSIONS AND POLICY IMPLICATIONS

The diseases acting simultaneously in Matlab population and causing a fatality situation prevailing in Matlab, may have important policy implications. The findings of our study may help planner and policy makers to take appropriate measures to reduce mortality levels in the study area, Matlab, as well as of the country, which faces a serious problem of health condition vis-à-vis socioeconomic underdevelopment and discouragement of high fertility of our population. Most mortality due to diarrhoeal diseases in almost all ages is reflected from the analysis. Diarrhoeal and infectious diseases seem to be the major causes of low life expectancy in Matlab as well as in our country. Therefore if we are able to eliminate or control diarrhoeal and infectious diseases from our community expectation of life will increase about 1.43 years at birth. It is apprehended that the problems are associated with controlling the diseases of pneumonia, diarrhea, dysentery, tetanus, small chicken pox and malnutrition in Matlab. Special attention should be given towards children with emphasis early of approximate dehydration therapy to minimize their mortality rates.

The strategy of universal health care by the year 2000 has become accepted, and government efforts toward infrastructure development included the widespread construction of rural hospitals, dispensaries, and clinics for outpatient care. Presumably, use of safe water has substantially reduced mortality and morbidity due to water born diseases. If the situation is improved, further control over this type of disease can be achieved. Infant and children less than five years of age died with greater intensity due to diarrhoeal diseases as well as some remarkable deaths due to infectious diseases. Though the study is meant for a subpopulation and thus not amenable for generalization, still the findings may prove of value to policy makers, planners and researchers in exploring the possibilities of improving health and hygienic conditions and controlling the most vulnerable diseases operating not only in Matlab but also in the whole of Bangladesh.

REFERENCE

Abedin, S. and Islam M. N. (1993) "Structure of Death of water related Infectious Diseases in a Rural Community of Bangladesh." Proceeding of the fourth National Statistical Conference on Population, Health and Environment in Bangladesh. January 15-16, 1993; pp 55-65.

Ajit Kumar Sarker (2005). Identity of population Ageing in Bangladesh: A Statistical investigation. Unpublished M. Sc. Thesis, April 2005.

Aung, T. (1984). ARI in children in Burma. ARI in childhood proceedings of an International Workshop, Sydney, Agust, 1984; 93-96

Barua, D. (1981). Global impacts of applied microbiology. Sixth International Conference. London, Academic Press 1981; 405-413.

BBS (1997). Statistical Year Book of Bangladesh, 1996

BBS (2003). Statistical Year Book of Bangladesh. 2002

BBS (2003). "Report on Sample Vital Registration System 1999-2002", Percent Distribution of death by causes.

Dutta, S. (1997). Isolation of Escherichia Coil to Detect Faecal Contamination of Infant and their Mothers in West Bengal. Journal of Diarrhoeal Diseases Research. Vol. 15, No. 3, 1997.

Elandt-Johnson R. C. and Johnson, N.L. (1980). Survival Models and Data Analysis. John Wiley and Sons.

George, et al. (1997). "Character's of knowledge and preparation abilities of salt sugar solution in North-Eastern Nigeria." Journal of diarrhoeal diseases research, Vol.15, No. 4, 1997.

Hoque, M. A. (2008): Disease Effect on Life Expectancy in Bangladesh: A Decrement Analysis. Bangladesh Journal of Environmental Science. Vol. 13 No. 2. PP. 260-267.

Hoque, M.A. (1993). Trends in mortality and probability patterns of death in a rural population of Bangladesh. Unpublished M. Sc. Thesis, December 1993.

Hoque, M. A. and Abedin, S. (1996). Old age Mortality in Matlab: A Multiple Decrement Analysis. Proceeding of the International Seminar on Ageing in SARCE Countries. 18-19 October 19-96: 92-98.

Hoque, M.A. and Islam M. N. (2002). Effect of elimination of a specific disease on life expectation of Bangladesh. Man in India, 82 (1& 2): pp. 155-163.

Kabir M, Chowdhury RI and Amin R. (1995). Infant and child mortality levels and trends in Bangladesh. J Biosoc Sci Apr; 27(2): 179-92.

Karn, M. N. (1933). A further study of constructing life tables when certain causes of death are eliminated. Biometrika, 25. 191-201.

Khan, F. (1986). How to prevent Tetanus from Injuries? Bangladesh Medical Review. Vol. XII, No. I & II.

Malik, A. (1979). Prevention of Cardiovascular Diseases. Bangladesh Medical Journal. Vol. 7, No. 3 & 4.

Ministry of Labor, Health and Social Affairs of Georgia (Third Edition, November 2005).
Surveillance and Control of Communicable Diseases: Guidelines for Public Health Services in Georgia.

Staffs of ICCDR, B (2005) "Health and Demographic Surveillance System (HDSS)- Matlab." Vol. No.36, Registration of Health and Demographic Events 2003. Scientific Report No. 92.

Staffs of ICCDR, B (2002) "Health and Demographic Surveillance System (HDSS)- Matlab." Vol. No.33, Registration of Health and Demographic Events 2000. Scientific Report No. 89.

Snyder, J.D. and Mersen, M.H. (1982). The Magnitude of the global problem of acute diarrhoeal disease: are view of active surveillance data. Bull. WHO 1982; 60;

The Bangladesh Observer (2005).Unthinkable Demographic Changes: Fertility, mortality rates decline. 29-12-2005.

The Bangladesh Today (the daily newspaper of Bangladesh); Fri. March 24, 2006: "TB Claims 1.7 Million Live in 2004."

WHO/UNDP (1986) "Basic Principle for Control of ARI in Children in Developing Countries ARI in South East Asia." SEARO Technical Publication No.8, Delhi1986, pp. 39-48.

WHO (1986) ARI/1, Geneva; WHO 1-2.

World Health Condition (2005) "Gordon, David Promukh: Child Poverty in the Development World, The Policy Press, Bristol, UK, October -2003."

UNDP; ESSENTIALS NO. 6 June 2002: "A Multifaceted Response to the HIV/AIDS Pandemic."