Development of indicators for occupational health and safety surveillance:
The Thai and Pan American Health Organization experiences

Introduction

Occupational health and safety surveillance is a key step in the promotion of health and safety at the workplace. Surveillance is defined as the "ongoing systematic collection, analysis, and interpretation of health data essential to the planning, implementation, and evaluation of public health practice, closely integrated with the timely dissemination of these data to those who need to know (1)." To achieve the goal of promoting occupational health and safety, however, we do not need to collect, analyse, and interpret every single item of data at the workplace. In other words, surveillance calls for an indicator approach (2). The question then becomes: How do we select indicators for occupational health and safety surveillance?

This paper addresses two of my experiences with indicator development for occupational health and safety surveillance, one in Thailand and the other for the Pan American Health Organization (PAHO).

The Thai Experience

In 1995 I was hired as a "Social Security Expert" by the International Labour Organisation (ILO). The place of recruitment was Toronto (Canada), the duration of appointment was two months, and the duty station was Bangkok (Thailand). The mission was to advise the Social Security Office of the Ministry of Labour and Social Welfare in Thailand on how to establish a surveillance system on occupational diseases and injuries under the Workmen's Compensation Fund (WCF). The rates of occupational diseases and injuries in Thailand are increasing. It was envisaged that, by developing the surveillance system, more knowledge about the causes of these occupational diseases and injuries could be obtained, thus contributing to the design and implementation of disease and injury prevention measures.

The list of my duties included, among other things, identifying the information needs of the WCF, specifying data elements to be collected in order to meet the information needs, and outlining the analyses needed to convert the data into information. In other words, two sets of indicators needed to be developed during my two-month assignment: information indicators and statistics indicators. What kinds of standard indicator information need to be collected in the notification forms? What kinds of standard indicator statistics need to be produced in the WCF annual reports?

I first spent a week at the ILO Headquarters in Geneva, to be briefed for the assignment. The first two days in Thailand were fully occupied by briefing at the ILO Regional Office in Bangkok, and by a joint meeting of the Social Security Office (SSO), WCF, ILO, and the National Institute for the Improvement of Working Conditions and Environment (NICE).

During the two-month assignment, a series of tasks was accomplished in order to develop indicators for the surveillance system. First, a comprehensive literature research was conducted. Second, a review of key literature was performed, including the ILO Code of Practice (3) and other international recommendations on statistics on occupational accidents and diseases (4,5). Third, a comparative study of the indicator information collected by various notification systems in Australia, Canada, the Philippines, Sweden, Thailand, and the United States was carried out. Fourth, fact-finding sessions were held with the information suppliers and users in Thailand. Fifth, based on selected information indicators, the existing Thai notification form was redesigned and then pretested. Sixth, coding schemes for the information collected on the redesigned form were constructed. Seventh, a list of statistics indicators was formulated for inclusion in the WCF annual report. Lastly, suggestions were made for future improvements of the proposed surveillance system. Details of the ILO assignment have been documented elsewhere (6).

A number of things were learnt from the Thai experience. In view of the facts that a surveillance system takes time to implement, and that a large number of records need to be entered, the proposed list of information and statistics indicators consisted of a minimum required set (to be implemented immediately), and a more detailed set (to be implemented when the capacity of the Thai system allows). This approach might at least partially solve the problem posed by Malison, who claimed that "one of the most prevalent characteristics of surveillance systems in the developing world is that they attempt to collect too much information about too many diseases and conditions (7)."

Indicators are culture-specific. For example, "sex" (gender) and "marital status" are considered sensitive questions in Thai culture and are not asked in the notification form. Instead, the required information is inferred from the response to "title" (Mr/Mrs/Ms). As another example, the standard indicator questions used in Sweden on the level of safety provided by the employer, "Any supervision?", "Any safeguard?", "Any safety equipment?", and "Any training?", cannot be used in Thailand. Owing to the workers's low level of literacy, most occupational injury and disease notification forms are filled out by the employer (personnel department staff); the result is that all of these questions are answered "YES".

Interpretation of indicators, too, is also culture-specific. The standard coding scheme for occupation, recommended by the ILO Code of Practice (3), cannot be used in Thailand. The ILO scheme does not have certain occupations that are specific to Thailand, such as crocodile farmers.

The PAHO Experience

On 7-9 July 1999, I participated as a Canadian expert in three-day workshop organized by PAHO, chaired by Dr. Maritza Tennassee and Dr. Gerry Eijkemans and held in Washington, DC (USA). (For a recent profile of Dr. Tennassee, please see the August 1999 issue of the Asian-Pacific Newsletter (8).) This workshop utilized an approach to indicator development which emphasized the importance of international collaboration. A total of 24 occupational health experts from 13 countries in the Americas were invited.

An objective of the workshop was to improve the quality of data on workers' health and safety in the Americas. More specifically, the experts were asked to develop and prioritize a list of possible occupational sentinel health events (indicators) for surveillance.

Participants were divided into three groups, each of which independently discussed and analysed the occupational conditions (health events, hazards) considered to be a priority for surveillance in workers' health. The three groups were asked to present a list of five to ten conditions, based on the following criteria: Magnitude; severity; costs; and intervention/prevention possibilities. In addition, the groups were asked to develop their own scoring systems. For example, the first group used low (1 point), medium (2 points) and high (3 points) categories to rate the four criteria (magnitude, severity, costs, and intervention/prevention possibilities) together, and each member of the group contributed an individual score. The second group assigned a group consensus score of +, ++ and +++ - indicating priority from low to high - to each of the four criteria separately. The third group used group scores of 1 to 3 (low to high) for each of the four criteria (similar to the second group but not the first). These scoring systems were different, but the results of the prioritization of indicators among the three groups were very similar (9).

A number of things were learnt from the PAHO experience. First, the process of expert consensus is an efficient method for the development of indicators for a surveillance system. Second, scoring systems can be used to register the level of consensus among the experts and to prioritize indicators. Third, the actual method of scoring does not seem to affect results with regard to prioritization of indicators.

Discussion

In the development of indicators for occupational health and safety surveillance, two important underlying principles are: the dissemination of information; and international collaboration. Feedback and the dissemination of information is an important step in surveillance (10), and should be done right from the start. In the Thai project, three joint meetings of the SSO, WCF, ILO, and NICE were held throughout the two-month assignment, to inform the stakeholders of the progress and to obtain suggestions. In addition, all information suppliers and users were consulted through fact-finding sessions and feed-back comments on progress and final reports (6).

The importance of international collaboration is well illustrated in the PAHO project. There were three expert groups: two Spanish-speaking groups and one English-speaking group. Plenary sessions were held with simultaneous interpretation. Language, culture, or geographical distance should not be a barrier to international collaboration. As more different viewpoints are presented, problems are more likely to be resolved successfully.

References

1. Centers for Disease Control. CDC surveillance update, Atlanta (Georgia): CDC, 1988.
2. Choi BCK. Perspectives on epidemiologic surveillance in the 21st century. Chron Dis Can 1998;19:145-51.
3. International Labour Office. Code of practice on recording and notification of occupational accidents and diseases. Geneva: ILO, 1994.
4. International Labour Office. Year Book of Labour Statistics, 1994. 53rd issue. Geneva: ILO, 1994.
5. International Labour Office. Current international recommendations on labour statistics. Geneva: ILO, 1988.
6. Choi BCK. Recording, notification, compilation, and classification of statistics of occupational accidents and diseases: The Thai experience. J Occup Environ Med 1996;38:1151-60.
7. Malison MD. Chapter 5. Surveillance in developing countries. In: Halperin W, Baker EL, Jr (eds). Public health surveillance. New York: Van Nostrand Reinhold, 1992:56-61.
8. Lehtinen S. Women in the world of work. Asian-Pacific Newslett Occup Health Safety 1999;6:44-5.
9. Choi BCK, Eijkemans GJM, Tennassee LM. Prioritization of occupational sentinel health events for workplace health and hazard surveillance: The PAHO experience, in preparation, 2000.
10. Klaucke DN. Chapter 3. Evaluating public health surveillance systems. In: Halperin W, Baker EL, Jr (eds). Public health surveillance. New York: Van Nostrand Reinhold, 1992:26-41.

Dr. Bernard C.K. Choi
Chief Epidemiologist
Bureau of Cardio-Respiratory Diseases and Diabetes
Laboratory Centre for Disease Control, Health Canada, AL#1918C3, Tunney's Pasture
Ottawa, Ontario, Canada K1A 0K9
E-mail: Bernard_Choi@hc-sc.gc.ca

Dr. Choi is also an Associate Professor at the University of Toronto and an Adjunct Professor at the University of Ottawa.