[ Table of contents ]   [ List of papers ] Health Impact of Occupational Risks in the Informal Sector in Zimbabwe Chapter 5 Future surveys of occupational risks and their health impacts

The strong possibility indicated of undertection and under reporting of occupational illness and injury across all sectors highlights the need for a more comprehensive survey of occupational risks and their health impact. The sample survey of the informal sector reported here indicated particular exposures and health outcomes that contribute to a large share of occupational morbidity, and further suggested that the burden of morbidity is significantly greater than estimates obtained from current reporting systems. These additional burdens may appear in the public health systems, but are not recorded as occupational in origin.

The lack of a systematic assessment of occupational health risks, practice and outcomes makes it difficult to assess how economic and production policies are affecting occupational health, the costs of such impacts and the potential benefits to be gained from investments in prevention. It also makes it difficult for policy makers or those who allocate resources to perceive the real gains from resource allocations towards occupational health practice, or to assess the performance of their investments in this area. Without systematic indicative information, occupational health outcomes are often not incorporated as factors in production decisions, except in as far as the regulatory framework requires it, in as far as producers perceive that such regulation will be enforced and that penalties for non enforcement will be a deterrent. With the low levels of penalties currently applying^{(Endnote 11)} and the weakness of state inspection systems, it is evident that regulation is a necessary but not sufficient mechanism for meaningful investments in occupational health practice.

Some countries have included occupational injury/illness into national household surveys and to develop surveillance systems for occupational morbidity. In USA, UK and South Africa, surveillance systems have been established for the surveillance of the most common occupational illnesses, particularly respiratory disorders and skin disorders. These systems operate at their base in the first level contact with general medical practitioners (private and public), and link with specialist occupational medicine training, referral and support. They have been used to provide early notification of clusters of occupational illness and to trigger prevention of further disease, but have also been used to provide a national profile of the distribution and possible determinants of such disease.

Other countries have included a module on occupational risk and morbidity in their national household survey programmes. Such countries, including UK, South Africa and Pakistan have developed questions that are used to elaborate the profile of occupational morbidity, its major causes and consequences, using both definitions enshrined in law and other definitions that are becoming more internationally standardised through ILO. The results have also been used in the UK for example to assess the costs of occupational morbidity and the distribution of these costs.

A second method is applied in Sweden, and consists of regular household surveys of reported exposures on selected important categories of risk, termed the 'National Survey of Work Related Problems'. Such surveys are carried out annually on a random sample of about 25 000 workers and are aimed at obtaining reported information on work environments, risk perception and health impacts. When compared with objective sources of exposure and health impact information the surveys have been found to provide a reasonably valid picture of the major risks and health impacts in sectors, and a sensitive assessment of changes in musculoskeletal and psychosocial risks (Hogstedt 1997).

In such surveys reported risks could incorporate certain sources of bias. Hazards not easily detected through sensory perception, such as certain dusts and chemicals, are likely to be underestimated. Risk perception has been found, through the Swedish surveys themselves, to relate to respondents' knowledge and expectations (Loewenson, Laurell and Hogstedt 1995). The reluctance to talk about illness and the lack of widespread perception that work may cause ill health may lead to significant underestimates of real levels of occupational illness, even when this is at clinical levels.

Carrying out an occupational health component of a national household survey would however begin to address the most significant source of bias at present, which is the almost complete exclusion of the majority of workers outside the formal sector of employment, and the significant underdetection of occupational disease, as distinct to injury. It may also enable the assessment of occupational morbidity and mortality arising in non employed populations who are also exposed to such occupational risks, such as the spill-over of occupational risks to people living in vicinity of farm, mine and industrial production, exposure through washing of work clothes and through air and water pollution (Packard 1989; Loewenson et al 1991).

The questionnaire used in the survey reported here acts as a useful starting point for the development of a module on occupational risks and outcomes in a household survey programme. It would need to be further discussed by relevant expertise, but some immediate issues arising from its administration in this pilot exercise are:

1. It would be useful to obtain information on last job (work activity) prior to current job and number of years in current work activity, both as an indicator of movement between sectors of the economy and also of the extent to which mixed exposures due to shifts between different types of work.
2. The question on reported number of people in the workplace and its later use as denominator data for occupational mortality estimates and the later question on reported occupational deaths in the past five years would need to be reviewed to assess what statistical adjustments should be done to make it a more valid tool. As used in this survey it was a rather crude measure. Such adjustments would need to take into account the possible over-representation of data from large workplaces where both numerator and denominator data would be inflated relative to that from small workplaces. This was less important in the informal sector survey, given the consistently small workplace size but would become important where much larger workplaces are included.
3. Some questions may not be relevant to a wider survey. On job status it is suggested that the questions on age, sex, education, main work activity in the past 12 months, (previous work activity noted in (ii) above), sector of employment, skills level, registration status of the workplace and number of people working in the establishment were useful questions. On workplace exposures it is suggested that a smaller number (about 10-15) of the most common and severe hazards are included.

   Using the feedback from this survey and from prior formal sector surveys these could be

   • inadequate clean water supplies;
   • toilets inadequate / dirty
   • tripping hazards;
   • machine parts unguarded;
   • work with cutting/ grinding handtools
   • noise too loud to talk normally;
   • exposed to extreme heat source
   • use red or purple triangle chemicals
   • mineral dusts in workplace
   • vegetable dusts in workplace
   • handling heavy loads
   • uncomfortable work posture
   • long hours of standing

   On occupational health outcomes, the questions that appeared most useful from the survey reported here were on perceived dangers and the harm done, total accidents caused by work with more than 24 hours lost time and total accidents caused by work requiring a medical consultation; injury in past 12 months leading to permanent disability (with and without job consequences); injury reporting to and compensation from NSSA; work stoppage due to illness caused by work in past 12 months, specific health problems noted; reported number of workplace deaths due to accidents in the past 5 years.

   A further question on total lost work time due to work related injury and illness in the past year could be usefully included to assess the impact of such morbidity.

   It is however suggested that some amendments to the wording of these questions be made^{(Endnote 12)}.

4. The frequency of injury and illness assessed in the survey can be linked to other components of the household survey questionnaire, such as income and expenditure, to make estimates of cost burdens.

A national survey of occupational exposures and morbidity would begin to more clearly identify from a less biased data source where attention should be directed to reduce the major causes of morbidity. Periodic collection of such data would provide policy makers with estimates of the burden and distribution of the health impacts of occupational risk, the progress towards reduction of that risk and the returns to their investments in occupational health practice. It is suggested that this inclusion of work related health outcomes in the national systematic assessment of public health in Zimbabwe is long overdue.

Endnote 11:
At present the maximum penalty for breach of the law is Z$2 000

Endnote 12:
These will be included after the meeting with the field workers has been held.

[ Table of contents ]

[ List of papers ]

International Labour Organization (ILO): Contact us | Site map |

Copyright and Permissions 1996-2008 International Labour Organization (ILO) - Disclaimer