Impact of HIV infection on Zambian businesses
BMJ 1994; 309 doi: https://doi.org/10.1136/bmj.309.6968.1549 (Published 10 December 1994) Cite this as: BMJ 1994;309:1549
- Rachel Baggaley, medical and research coordinatora,
- Peter Godfrey-Faussett, lecturerb,
- Roland Msiska, director Zambart projecta,
- Diane Chilangwa, research assistanta,
- Eusabio Chitu, HIV outreach programme coordinatora,
- John Porter, senior lecturerc,
- Michael Kelly, director National AIDS Control Programme, Ministry of Health, Zambiaa
- a Kara Counselling and Training Trust, PO Box 37559, Lusaka, Zambia
- b Department of Medicine, PO Box 50110, University Teaching Hospital, Lusaka, Zambia
- c London School of Hygiene and Tropical Medicine, London WC1E 7HT
- Correspondence to: Dr Baggaley.
- Accepted 30 September 1994
Women attending antenatal clinics in Zambia have rates of HIV infection of 11-30%.1 Deaths from the disease are likely to affect the economy of individual families and, if widespread, that of the country. Since December 1990 the Kara Counselling and Training Trust has offered education about HIV to local companies. We therefore studied the impact of HIV infection on businesses in Zambia as reported by senior management staff.
Methods
and results
A questionnaire designed to assess the quantitative and qualitative impact of HIV infection on businesses was delivered to the personnel managers of 33 companies, including banks, car manufacturers, food and drink manufacturers, and farms. One week later it was collected and any ambiguities were clarified. Death rates were calculated by dividing the annual number of deaths by the total number of employees at the end of each year. Mortality and morbidity trends over the past six years were studied by χ2 analysis.
All 33 companies filled in questionnaires: 25 were based in Lusaka, the capital (population 982 000, 1990 census) and eight in the Copperbelt, the other main conurbation (total population 1 200 000, 1990 census). Thirty two questionnaires were returned fully completed. One company could not find its mortality records. The total number of employees in 1993 was 10204 (79% men, 21% women); the range for individual companies was 19 to 1836. The total number of employees did not change significantly over the study period. Most companies had a workforce aged between 18 and 55 years. Four companies had employees over 55 and one over 65.
The crude death rate in this population increased sequentially (P<0.001, χ2 test for trend) from 0.25 per 100 person years in 1987 to 1.83 per 100 person years in 1993 (table).
Comment
In the developing world average annual mortality between the ages of 15 and 60 is estimated to be around 0.5 per 100 person years.2 Vital registration systems are rare in developing countries, but in Zambia company mortality records are likely to be accurate because funeral benefits are substantial. Some employees could have retired early because of ill health and so would not appear in the mortality figures when they died. We believe, however, that few deaths will have been missed because normal employment practice allows employees to remain on sick leave for prolonged periods. Buve et al found that mortality in female nurses from two hospitals in Zambia rose from 0.2-2.7 per 100 person years between 1980 and 1991.3 Our data, collected from a much larger population, fall well within their 95% confidence interval.
Crude annual mortality data and causes of death among employees of 33 Zambian businesses from 1987 to 1993
This increase in crude mortality cannot be definitely attributed to HIV infection as most companies do not record the cause of death. There was an increase in reporting of death from HIV infection and also from tuberculosis and diarrhoea, which are strongly associated with HIV infection in Africa.4 5 HIV testing has not been widely available in Zambia and great stigma still exists. Deaths from HIV infection are seldom recorded, even in official statistics, since both doctors and relatives are reluctant to acknowledge publicity that HIV was the cause.
The businesses were not selected randomly. Some had approached Kara Counselling to ask for HIV education programmes while others had been approached through an informal network of contacts in the business community. Unemployment is a major problem in Zambia. The study population therefore does not represent those with the least income and the worst nutrition.
The increase is sufficiently striking to make it unlikely that changes in the demography of the study population could account for it. None of the managers interviewed commented on such changes. Managers in some of the larger companies had noticed a rising mortality among their own workforces, but the aggregation of various different businesses provides strong evidence that the death rate is rising among Zambian employees.