Elaboration and validity of an instrument for assessing good handling practices for fishmongers to guide the sanitary surveillance in Rio Grande do Sul/Brazil

Fish consumption has increased in recent years due to the increased supply and awareness of the health benefits of fish proteins. Fresh fish and its products can be purchased in fish retail trades known as fishmongers. However, there is no specific regulation in the state of Rio Grande do Sul (RS)/Brazil to provide a checklist for the inspection of fishmongers. Thus, this study aimed to elaborate and validate a complete easy-to-use checklist of good handling practices for fishmongers (LVBPMP) to assist municipal sanitary inspectors of these establishments. For that, 11 health inspection scripts focused on fishmongers were consulted, concerning municipal, state, federal, and international regulations, resulting in 1,188 items analyzed. Of these, 74 items were selected and divided into six categories according to current Brazilian legislation. A pre-test was carried out on two fishmongers in Porto Alegre/RS, and then the LVBPMP was analyzed by a committee of six experts for the face validity of the instrument. The validated LVBPMP comprised 84 items. LVBPM proved to be a complete and objective instrument to guide the sanitary surveillance staff in the inspection of fishmongers, allowing its use by regulatory agencies to assist the municipal sanitary inspectors of fishmongers in the RS state.


Introduction
According to The State of World Fisheries and Aquaculture (SOFIA) of 2020, published by the Food and Agriculture Organization of the United Nations (FAO), global fish consumption increased at an average annual rate of 3.1% from 1961 to 2017, which is almost double the population growth (1.6%) in the same period, and higher than the consumption of other foods of animal origin (meat, milk, and derivatives), with a 2.1% annual increase (FAO, 2020).
The per capita consumption of fish in Brazil was 10.5 kg per inhabitant per year in 2021, which is below the FAO recommendation of 12 kg (PeixeBR, 2020). As reported by Santos (2022), this scenario has increased after three consecutive years of declined consumption of this protein. Sartori and Amancio (2012) reported that fish is seen as a food of great importance for human nutrition, mainly due to its nutritional value. It is considered a source of high biological value proteins and presents a balance of essential amino acids and fat-soluble vitamins, besides containing minerals such as calcium, phosphorus, and iron (Sartori & Amancio, 2012). In addition, it contains all essential amino acids, with high lysine content and the minimum amount of connective tissue, which favors digestibility above 95%, depending on the fish species, which is higher than that of meat and milk (Oetterer et al., 2006).
The World Health Organization (WHO) recommends the consumption of at least two portions of some variety of fish products a week to supplement eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are essential to the human body due to their cardioprotective effects (Hellberg et al., 2012).
The commercialization of fish is carried out in fish retail stores called fishmongers. In Brazil, this economic activity is included in the National Code of Business Activity (CNAE) under registration number 4722-9/02, which includes the retail trade of fresh, frozen, preserved, or refrigerated fish, crustaceans, and mollusks (Concla, 2021).
The municipal health surveillance authorities are responsible for the inspection of fishmongers in the municipalities, aiming to verify the hygiene and sanitary conditions of the facilities and equipment, the health of the handlers, as well as the quality and safety of the stored, processed, and marketed fish.
Although sanitary regulations have been used to guarantee the adequate inspection of the fish market, there are no specific rules in the state of Rio Grande do Sul for this branch of activity, which impairs an effective sanitary inspection. In addition, fish is a highly perishable and easily degradable food, with high risk to the consumer when handled and marketed without the minimum requirements for hygiene and food control.
The FAO and WHO Joint Sanitary Code Commission has outlined preventive measures for the preparation, handling, storage, transport, and market for producing safe and suitable food for human consumption (WHO, 1968apud Marins & Tancredi, 2014. The Integrated Manual of Surveillance, Prevention, and Control of Food borne Diseases (Brasil, 2010) defines safe food as those products manufactured, preserved, transported, sold, or exposed for sale, under conditions that ensure the control of hazards, without risk to the consumer.
According to the World Health Organization (WHO), food borne diseases encompass diseases of the broadest spectrum and result from the ingestion of food contaminated with microorganisms or chemicals, being considered a growing public health problem (Brasil, 2010). Huss et al (1997) states that diseases transmitted by fish consumption may be associated with contamination by microorganisms such as Clostridium perfringens, Escherichia coli, Vibrium cholerae, among others. In addition, the presence of chemicals, such as marine biotoxins, are also associated with the consumption of contaminated fish.
Considering the importance of fish consumption, the lack of specific sanitary legislation for fishmongers in the state of Rio Grande do Sul/Brazil, and the food safety threat of this product, this study aimed to develop and validate a checklist of good handling practices for fishmongers that is easy to apply, up-to-date, and objective, thus contributing with the municipal sanitary surveillance team of Rio Grande do Sul during inspection of these establishments. Research, Society andDevelopment, v. 11, n. 12, e456111234538, 2022 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v11i12.34538 4 The flow chart in Figure 1 demonstrates the experimental design of the process of elaboration and validation of a handling checklist for fishmongers (LVBPMP), which was composed of the following steps: literature review, in which eleven best practice checklist for fishmongers from different administrative spheres were evaluated (municipal, state, federal and international); preparation of the checklist handling practices for fishmongers (LVBPMP); classification of the items according to heath risk in Indispensable, Necessary and Recomendable. From these initial stages, the Version 1 of the LVBPMP was elaborated, with 74 items, which was applied, as a pre-test, in two fishmongers. For validation of the experiment, face validity was applied by 6 experts, in two rounds, after the elaboration of the first and second versions of the LVBPMP. The Final Version of the instrument was written with 84 items.

Elaboration of a checklist of good handling practices for fishmongers
A checklist of good handling practices for fishmongers was developed from a literature review and analysis of existing instruments for the evaluation of hygienic-sanitary conditions of commercial establishments, with a focus on fish that included storage, processing (peeling, skinning, gutting, and filleting) and commercialization of fish as described in the State To structure and organize the LVBPMP, the categories of the good manufacturing practices checklist of ANVISA (Resolution RDC 275, of October 21, 2002) were used, with adaptations. Although it was not a specific list for fishmongers, it was considered the most appropriate basis due to its scope, wide dissemination, excellence, and wide use by sanitary inspection.
The categories of the ANVISA checklist (Resolution RDC 275, October 21, 2002) are classified according to the standard operating procedures (SOPs), which are determinants for implementing good practices in food service establishments. Therefore, the checklist proposed and adapted for the present study was organized into six categories, as follows: i) identification of the fish monger; ii) documentation; iii) buildings and facilities; iv) equipment, furniture, and utensils; v) handlers; and vi) raw material.
The first checklist contained 74 items (Version 1 of the LVBPMP) that covered most of the evaluation steps for the good handling practices and hygienic-sanitary conditions in a fish monger. Of the total of 74 items, 65 were objective, consisting of the responses Yes -Y (meets the requirements of the evaluation item), No -N (does not meet the requirements of the evaluation item), and Not Applied -NA (the evaluation item does not apply to the inspected establishment or was not observed during the application of the instrument). Research, Society and Development, v. 11, n. 12, e456111234538, 2022 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v11i12.34538 5 All 74 items on the checklist were also classified according to the guidelines of the Ministry of Health (Brasil, 2010) into Indispensable (I), Necessary (N), and Recommendable (R). The classification criteria were based on the potential risk inherent in each item, concerning product quality and safety, and worker safety in their interaction with products and processes (Brasil, 2010). The indispensable items (I) are those that meet the good manufacturing practices and can greatly affect the quality or safety of products and workers in their interaction with products and processes. The necessary items (N) are those that meet the good manufacturing practices and can influence to a less extent the quality or safety of products and workers in their interaction with the products and processes. Finally, the recommendable items (R) are those that meet the good manufacturing practices and may have little effect on the quality or safety of products and the safety of workers in their interaction with products and processes (Brasil, 2010).

Pre-test in the fishmongers
A pre-test using Version 1 of the LVBPMP was performed on two fishmongers in the city of Porto Alegre/RS. The application of the checklist was performed by the researcher of this study, accompanied by an inspector from the food surveillance team of the General Directorate of Health Surveillance (DGVS) of the city of Porto Alegre.

Validity of LVBPMP
The apparent or face validity method was used to validate the LVBPMP, as proposed by Pasquali (2010). In this method, the evaluation instrument is analyzed by a committee of experts regarding the measurement capacity, measurement of effectiveness, and the relevant evidence, in addition to evaluating the structure and appearance of the instrument. The expert team was made up of health or agricultural inspectors. The validity process of the LVBPMP was carried out in stages, using Version 1 and Version 2 of the document, until a validated Final Version for use in a larger number of fish mongers.
For the analysis, six previously selected experts who agreed to participate in the study received Version 1 of the LVBPMP via e-mail, followed by Version 2 after four months. The experts were asked to evaluate and suggest changes in the instrument that they considered pertinent and necessary for its improvement.
The experts were selected as suggested by Oliveira et al (2014), as follows: i) academic background (multidisciplinary, with knowledge in the food area and good practices) in veterinary medicine, food engineering, nutrition, biology, or related fields; ii) position of health or agricultural inspector; iii) appointment by public competition (municipal, state, or federal spheres), and iv) experience in inspection of fish and fish mongers. The experts also received a link to Google Forms to evaluate the checklist. The methodology Seven Dimensions of Evaluation Quality proposed by Equal (2007 apud Stedefeldt et al. 2013) was used in the questionnaire, with the following evaluation criteria: i) contemplation and innovation, ii) benefit, iii) adequacy, iv) utility, v) accessibility, vi) equality, and vii) transference. The experts assigned a score for each question according to the five-point Likert Attitude Scale (Likert, 1932apud Stedefeldt et al. 2013), corresponding to 1strongly disagree, 2 -partially disagree, 3 -neither agree nor disagree, 4 -partially agree, and 5 -strongly agree.

Statistical analysis
All guidelines of the experimental design, the methodologies for validation, and the statistical analyses were

Pre-test
A pre-test was performed on two fishmongers in Porto Alegre, named "A" and "B", to determine the applicability and objectivity of the LVBPMP. Both fishmongers were intentionally selected to apply the instrument to fishmongers with different characteristics.
The fishmonger "A" was located in the northern of Porto Alegre/RS, with an average area of 60 m2, and had only one employee and no technical manager. Mullet fish was the main fish commercialized in the establishment, and the average monthly sales in volume were approximately 200 kg. The establishment commercialized frozen fish coming directly from fish distributors, with sanitary certification of origin and purchase invoices. It also bought fresh fish from fishermen from the northern coast of Rio Grande do Sul, and the freezing steps were performed at the fishmonger, with no invoice, inspection record, or any other form of sanitary certification of origin.
In turn, fishmonger "B" was also located in the northern of Porto Alegre/RS, inside a supermarket. It had an area of 30 m2, 4 employees, a technical manager, and Saint Peter (a variety of tilapia) was the main fish marketed in the establishment, with average monthly sales in volume of 4 tons. Fish were marketed as fresh, chilled, and frozen, and the latter was placed directly on the shelves, without handling. The results of the pre-test are shown in Table 1. Due to the unsatisfactory hygienic-sanitary conditions, represented by only 20% of compliant items in fishmonger "A", the establishment was notified by the municipal health surveillance of Porto Alegre to correct the irregularities within the period established by the health inspector. The action was considered appropriate, once most of the nonconforming items were classified as indispensable and necessary, as can be seen in Table 1, thus posing a high risk of an outbreak of food borne illness.
The fishmonger "B" had 90.77% of items in compliance with the checklist, thus meeting the sanitary and hygienic criteria. In addition, the number of indispensable items in compliance (22) also characterized fishmonger "B" as having good sanitary conditions (Table 1).
The results of the pre-test showed that the proposed LVBPMP instrument was easily understood and applicable by the sanitary inspectors, once it adapted well to the very different fishmongers and was considered complete and objective for all situations, allowing a better technical basis for decision-making in notifications by the sanitary inspectors.

Validity of LVBPMP
In the first round of evaluation of the good handling practices checklist by the experts (Version 1 of the LVBPMP), the following results were verified regarding the seven dimensions of evaluation (Table 2). The graphical representation of the average scores assigned by each expert for each dimension in Table 2 is shown in  Source: Authors.
The scores in Figure 2 showed that the dimension equality received the higher scores, with a final average of 5.00, demonstrating that the LVBPMP can be used in all types of fish mongers, as long as the 4722-9/02 code is provided by the CNAE. However, adequacy had the lowest score (3.83), mainly due to the scores assigned by the agricultural inspectors (experts D, E, and F), probably because these professionals are veterinarians working in inspection rather than fish retailing.
They reported that the checklist of the new version of the LVBPMP should contemplate items concerning the organoleptic characteristics and the sanitary inspection of fish.
On the other hand, the health inspectors (experts A, B, and C), consisting of two veterinarians and a nutritionist, all with experience in the fish market, gave higher scores for all dimensions, except for accessibility, in which the nutritionist (expert C) recommended a clearer and more direct description of the items in the new version of the LVBPMP.
In addition to answering the questionnaire, the experts also suggested changing 28 items in the checklist, including the classification and/or writing of the items, inclusion and/or exclusion of items, improvement of items described with negative sentences, and instructions for completion at the end of the checklist.
All suggestions of the six experts were taken into account, thus Version 2 of the LVBPMP was expanded to 80 items.
The results of the second round of evaluation (Version 2 of the LVBPMP) for the seven dimensions of evaluation are shown in Table 3. The graphical representation of the average scores assigned by each expert for each dimension in Table 3 is shown in  Source: Authors.
The dimensions benefit and accessibility showed the highest scores (4.75), followed by adequacy, utility, equality, and transference (4.67), and innovation and contemplation with an average score of 4.50, as can be seen in Figure 3.
Considering the two rounds of evaluation, all dimensions had final scores above four, representing a good evaluation of the instrument by the experts. Table 4 shows the average scores of the first and second rounds and the final scores of each dimension. As can be seen in Table 4, the dimensions equality, transference, and accessibility presented the best scores in both rounds of evaluation, thus the LVBPMP has proven to be adequate, with the potential to improve the inspection processes of the local sanitary surveillance, making them more agile and focused on the final decisions. It also indicated that the instrument had clear and objective questions, was easy to understand and interpret, and can be easily integrated into professional inspection practice.
The experts suggested changes in the structure of Version 2, with the inclusion of six new items and the exclusion of two items, due to ambiguity of information. After the changes, the Final Validated Version of the LVBPMP contained a total of 84 items (Figure 4). A summary of the changes proposed by the experts in the first and second rounds of evaluation is shown in Table 5.  When assessing the items by health risk, the experts had the opportunity to modify the initial rating. The checklist of good handling practices for fish mongers was composed of 74 items, of which the first nine were not tabulated once they corresponded to the identification of the establishment, while the remaining 64 items were initially considered indispensable by 35.9% (23), necessary by 46.7% (30), and recommendable by 17.1% (11) of the experts. Table 6 shows the changes in the classification of items by health risk suggested by the experts, resulting in the finalized instrument, with 74 classifiable items.  Oliveira et al., (2014).
In the first round of evaluation, there was an increase in the number of items classified as indispensable, mainly in Category 6, which refers to the raw material (Table 6). The relevance of this category when compared to the other categories was confirmed in the second round of evaluation, which culminated in the final version of the LVBPMP. Thus, given the number of indispensable items in Category 6, during a sanitary inspection, the greatest attention of the inspector should turn to the items in this category.
As shown in Table 6, Category 3, which referred to building and facilities, had the highest percentage of items considered necessary. Thus, the noncompliance of these items can interfere to a less extent with the quality and sanitary safety of the marketed products, although they should meet the good manufacturing practices. Category 2 presented the highest number of recommendable items, probably for containing items that cover the documental part of the establishment, which are answered according to the information provided by the person in charge during the evaluation. Thus, an inspection action cannot be performed immediately in the case of negative responses once the failure does not lead to imminent health risk to the consumer. Research, Society andDevelopment, v. 11, n. 12, e456111234538, 2022 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v11i12.34538 13 Source: Authors.

Conclusion
The main topics required by the existing sanitary legislations applicable to fish mongers were contemplated in the final version of the LVBPMP. The indication of indispensable items in these establishments can contribute effectively to the results of sanitary inspections, aiming to enhance food safety.
The heterogeneity of the expert committee that validated the instrument contributed to a more differentiated view of fish mongers, from the quality and safety of the raw material to the supply to the consumer, allowing a complete evaluation of the instrument.
According to the validity rounds of the LVBPMP regarding the classification according to sanitary risk, Category 6 corresponding to the raw material should be considered the focus of sanitary inspections in fishmongers due to the largest number of indispensable items. However, further studies on the validity of checklists of GMP are required, which can contribute to the discussion topics.
Thus, the validated LVBPMP of the present study can effectively contribute to food sanitary surveillance actions in the Rio Grande do Sul by proposing a complete easy-to-use instrument to assist the sanitary inspectors during inspections in fish mongers, with the possibility of its adoption by regulatory agencies.