Hazard analysis is a confusing system for someone hearing about it for the first time, but imagine what it must be like when there is a language barrier as well. Shopna Begum and Andrew Mathieson investigate the effectiveness of the methods used to communicate and raise awareness of food hazards among proprietors whose first language is not English

With the increase in food scares and food poisoning outbreaks, the concern for hygiene standards in food premises is growing. Hazard analysis is a way of identifying potential hazards that may exist in the food preparation process and implementing control measures to minimise the risk of poisoning or contamination at critical control points in that process.(9)

The first language, and in many cases the only language, of the majority of EHOs in the UK is English. In many cases, when visiting a proprietor who has difficulty with the English language, the EHO has little choice but to leave translated material. In the case of achieving compliance with Regulation 4(3) of the Food Safety (General Food Hygiene) Regulations 1995 (as amended), the booklet produced by the Department of Health (DoH), A guide to food hazards and your business, is used. EHOs usually discuss the principles of hazard analysis before leaving any written material but this is not always possible. Often the best that can be achieved is explaining the complex theories to the proprietors' children in the hope that they will translate them effectively to their parents. This situation leaves some businesses relying entirely on translated written material to comply with complex legal requirements.

This research to determine the effectiveness of translated material was conducted on restaurant proprietors in Bristol, whose first language is Bengali. There are roughly 10,000 curry restaurants in the UK and Bengali is the first language of 68 per cent of such restaurant owners.(14)
Most of the Bangladeshis, perhaps as many as 95 per cent, came from the Sylhet district of Bangladesh. Their language is Sylheti, a dialect of Bengali, but differing sufficiently for a Sylheti speaker to have difficulty understanding the formal Bengali. Furthermore, Sylheti has no written form.(8,13)

METHODOLOGY
A hazard analysis worked example of a typical chicken dish was designed to help assess the level of understanding of the DoH's translated booklet. The example consisted of a multiple choice questionnaire that was easy to follow and quick to complete. It was translated into Bengali and proprietors were given the task of reading the DoH's booklet then completing the worked example.
A questionnaire was also designed to gather additional information on: the legal requirement; the DoH's booklet; training; the EHO; and personal details.
Examples and questionnaires were piloted and amended prior to collecting the main data set.
English-speaking proprietors (ESPs) were also surveyed as a control group so that a comparison between the effectiveness of the English and translated booklet could be made.
A telephone survey was also conducted on 20 randomly chosen local environmental health departments to discover what mechanisms were being used to overcome language problems.

RESULTS: The Worked Example

• Examples completed by respondents who have not had hazard analysis explained to them were all skewed towards the unsatisfactory category. Half of ESPs were categorised as permissible and 35 per cent were unsatisfactory. By contrast half of BSPs produced unsatisfactory results and 35 per cent were highly unsatisfactory.
• All of the stages within the worked example were critical to food safety but 40 per cent of ESPs and 73 per cent of BSPs failed to recognise this.

The legal requirement

• In total, 80 per cent of ESPs and 40 per cent of BSPs were aware of the legal requirement for hazard analysis. The majority had heard of the requirement through an EHO. Half of the aware ESPs had carried out hazard analysis, and 7 per cent of BSPs had done the same.
• 67 per cent of ESPs and 93 per cent of BSPs thought the written material was insufficient to understand and comply with the legal requirement.
• The proprietors from both communities preferred an EHO to explain hazard analysis in person rather than the booklet. However, 87 per cent of BSPs would only prefer this method if an interpreter could be used. A hazard analysis seminar and video was also deemed to be desirable.
• 40 per cent of BSPs said they would be reluctant to accept the Bengali booklet. They would prefer to have both the Bengali and English booklet so that one could be used to help the other.

The role of EHOs

• 73 per cent of the employees in ESP businesses were trained to the basic food hygiene level and 27 per cent had no formal training. All of the employees of BSP businesses were trained to the basic food hygiene level.
• All of the proprietors surveyed gave quite an accurate description of the role of EHOs in relation to their business.
• 20 per cent of ESPs and 67 per cent of BSPs have not had hazard analysis explained to them. All of these businesses had been visited by an EHO within the last year.

The telephone survey
All the local authorities surveyed were using the DoH's booklet to enable hazard analysis by business proprietors. However 90 per cent of departments thought the booklet was ineffective or were unsure about its effectiveness. The figure was 95 per cent for the translated booklet.

There were no ethnic minority EHOs at all in 35 per cent of the departments questioned. The number of ethnic minority EHOs in the remaining 65 per cent ranged from one to four, with the exception of 20 in one department. There were no Bengali-speaking EHOs.

The majority of worked examples completed by all those who were unaware of hazard analysis were not of a satisfactory standard. This indicates that the written training material used in isolation is ineffective. There was a slight difference in effectiveness between the English and translated booklet but it is difficult to say how significant this is.

There were respondents from both communities who failed to identify the cooling stage as critical to food safety, some proprietors also thought the reheating stage was not critical. This is worrying as it is almost impossible to avoid these stages in busy restaurants.

BSPs who were aware of hazard analysis were either fluent in English or had someone fluent in English working with them. Many BSPs said they would prefer an EHO to explain hazard analysis who was able to interpret the concepts into Bengali, otherwise they would prefer an interpreted seminar.

It must be noted that all those surveyed would like any interpretation to be done in Sylheti and not Bengali.
Some proprietors also expressed concerns that interpreters are not always themselves aware of what the exact legal requirements are and what various technical terms mean. This leads to misunderstandings and frustration. The obvious solution is to recruit EHOs with language skills. The other option is to train interpreters to understand the principles of hazard analysis, as well as what the law requires.

Another problem is that there are more employees than proprietors with difficulties with English. If the EHO explains in English or even if it is interpreted, the proprietor must still explain the principles to the employees who must then implement any requirements.

Usually at the end of translated material there is a number to call should there be any problems or if further information is required. However, worries about fluency and the availability of an interpreter may put proprietors off calling. This is another reason why the translated material is ineffective.

The telephone survey revealed a lack of ethnic minority EHOs in all local authorities. Local authorities say they try to use interpreters but it is not always possible due to financial constraints. It is evident that similar problems in other areas also exist as the same mechanism is being used by all local authorities. Departments were generally unaware that such a high percentage of food business proprietors speak Bengali as a first language. This shows a lack of data on the ethnic origins of proprietors in their areas which must be addressed if resources are to be effectively allocated.

CONCLUSIONS
The research indicates that both the translated and English booklet produced are ineffective when used in isolation. It also indicates that even when proprietors are aware of hazard analysis, they are still unsure of what is required of them.

The EHO explaining hazard analysis in person is the preferred method. Unfortunately this is not possible if there is a serious language barrier. There are very few bilingual EHOs and virtually no Bengali-speakers, and interpreters are expensive. If this continues then information will only be communicated to those fluent in English and local authorities will fail to provide a fair or effective service to the ethnic minorities running businesses within their districts.

The communication barrier must be tackled. If finance is tight, then seminars where information could be fed to a number of proprietors and staff at the same time should be considered. Even when an interpreter is used, they must be further trained so that they understand hazard analysis before they explain it to others.
Unfortunately, no matter how good an interpreter, the communication can never be as good as a direct one-to-one communication between the EHO and proprietors who share the same language and cultural background.

RECOMMENDATIONS
1. All EH departments should try to retain Bengali-speaking EHOs/interpreters and obtain more bilingual support literature in general.
2. Seminars should be organised to explain and demonstrate food hazard analysis to proprietors and staff whose first language is Bengali.
3. Any seminars for BSPs should be interpreted in Sylheti.
4. If interpreters are employed, local authorities must make sure that the interpreter understands the concepts and principles of hazard analysis.
5. All local authorities need to collect data to identify the ethnic composition of food business proprietors in their areas so that resources can be allocated accordingly.
6. A video needs to be produced to demonstrate hazard analysis in Sylheti for BSPs as well as other ethnic languages.
7. The DoH's booklet should only be used to aid discussion with proprietors regarding hazard analysis and not in isolation as it is ineffective when used in such a manner.
8. The difference between translated and interpreted must be understood and communicated widely to further better understanding.
9. Enforcement in premises where english is not the first language must be minimised until there is effective guidance, to the same standard as exists for those where English is spoken and understood.

Acknowledgements
The authors would like to thank Robert Lambourne (director) and Manjeet Gill (divisional director) at Bristol City Council for their support. Also: Dave Thomas, senior EHO at Bristol, Roger Cartwright, senior lecturer in Environmental Health at UWE, Abdul Wahab, information officer at Bangladesh Association, Bristol, and all proprietors and EH departments who participated.

REFERENCES
1. Ackerley L (1997) "Hazard analysis: a flexible approach for EHOs and the food industry", Environmental Health Journal, CIEH, vol. 105/2
2. Baker P, Hussain Z and Saunders J (1991) Interpreters in public services, Jo Campling
3. DoH (1996) A guide to food hazard and your business, DoH, London
4. DoH (1995) The Food Safety Act 1990, COP No. 9: Food hygiene inspections, London
5. DoH (1995) The Food Safety (General Food Hygiene) Regulations 1995, SI 1995 No. 1763, HMSO, London
6. Grice S and Perez D (1998) "Raising the standard of local food enforcement", Environmental Health Journal, CIEH, vol. 106/03
7. Hatim B and Mason I (1990) Discourse and the translator, Longman, London
8. House of Commons, Home Affairs Committee (1987) Bangladeshis in Britain, vol. 1
9. King H (1997) "What is hazard analysis?" Tandoori Magazine, vol. 3, issue 3
10. LACOTS (1997) Food Safety Hazard Analysis: Guidance on securing compliance with hazard analysis requirement Reg 4(3), LACOTS, London
11. Macdonald D J and Engel D (1996) A guide to HACCP: hazard analysis for small businesses, Highfield, London
12. Mortimore S and Wallace C (1994) HACCP: a practical approach, Chapman and Hall
13. Peach C (1996) The ethnic minority population of great Britain, vol. 2, Office of National Statistics
14. Wahhab I, Managing Director of Tandoori Magazine. 1998

Shopna Begum is an environmental health officer at Bristol City Council
Andrew Mathieson is a senior lecturer in Environmental Health, at UWE, Bristol

This study was conducted as a final year research project for the BSc (Hons) Environmental Health course at the University of the West of England. The opinions expressed are those of the authors alone.