The fallout from the Chung Lian carcinogenic oil incident continues to ripple across Taiwan, particularly sparking debate over the decision to allow food products containing up to 20% of the tainted oil to remain on shelves. In response, National Taiwan University public health professor Chien Chang-Chuan posted on Facebook, criticizing the policy as ineffective for food safety control and likening it to 'pouring oil on the fire.' He urged Taiwan to learn from the European Union (EU) by establishing an independent risk assessment system separate from government influence to prevent similar food safety crises in the future.

Chien pointed out that in early July, the Food and Drug Administration (FDA) announced a policy stating that food products containing less than 20% of soybean oil exceeding benzo(a)pyrene limits do not need to be recalled. However, the lack of clear scientific and legal justification for this decision left experts confused and triggered widespread public criticism over the benzo(a)pyrene contamination issue in soybean oil.

By early August, after the government released the minutes of the so-called expert meeting, the public discovered that the policy allowing products with less than 20% contaminated soybean oil to remain on shelves was derived through an improper process in which manufacturers were present. The decision was based on a simplistic dilution calculation: '8.1 ppb × 20% = approximately 1.6 ppb,' which is below Taiwan’s legal limit of 2 ppb for benzo(a)pyrene in soybean oil.

Chien emphasized that Taiwan’s 2 ppb benzo(a)pyrene standard for soybean oil was adopted from the EU. The EU standard, in turn, was developed by the European Food Safety Authority (EFSA), an independent scientific body, through a rigorous four-step risk assessment process.

EFSA identifies benzo(a)pyrene as one of the most representative carcinogens among polycyclic aromatic hydrocarbons (PAHs), recognizing its genotoxicity and carcinogenicity. This constitutes the first (Hazard Identification) and second (Hazard Characterization) steps of risk assessment.

The third step, Exposure Assessment, involves integrating actual benzo(a)pyrene concentrations in food, dietary intake levels, and consumption frequency to estimate cumulative population exposure from various food sources.

The fourth and final step, Risk Characterization, acknowledges that genotoxic carcinogens theoretically have no safe exposure threshold. Therefore, EFSA does not establish an Acceptable Daily Intake (ADI) or Tolerable Daily Intake (TDI). Instead, it uses the Margin of Exposure (MOE) to assess public health risk: MOE = BMDL10 ÷ human exposure level. BMDL10 is the lower confidence limit of the benchmark dose causing a 10% increase in tumors in animal studies. When MOE < 10,000, the public health risk is considered unacceptable, and risk management measures (e.g., recalls) are recommended. When MOE ≥ 10,000, the risk is deemed low.

Chien stressed that under EFSA’s principles, food safety cannot be judged solely by the proportion of contaminated oil in a product (e.g., 20%). To accurately assess human exposure, actual benzo(a)pyrene levels in final food products—such as cookies, bread, instant noodle seasoning oil, and salad dressings—must be measured. Without such data, it is impossible to know whether the contamination is evenly distributed, whether batch variations exist, or whether localized high concentrations occur during processing.

Since genotoxic carcinogens like benzo(a)pyrene have no confirmed safe exposure level, the focus should not be on 'how much is dangerous,' but rather on 'how much are people actually exposed to' and whether that exposure results in an MOE above 10,000.

Chien criticized the Ministry of Health and Welfare for making the decision—'food containing oil with 8.1 ppb benzo(a)pyrene at less than 20% concentration is safe and does not require recall'—without completing actual food testing, dietary exposure assessment, MOE analysis, or uncertainty analysis. This bypasses the standard scientific risk assessment process and represents a flawed risk management judgment.

EFSA’s 2008 risk assessment report concluded that measuring benzo(a)pyrene alone is insufficient to reflect the full carcinogenic PAH contamination in food. Some foods with undetectable benzo(a)pyrene still contain other carcinogenic PAHs. Therefore, EFSA recommends monitoring the total concentration of four PAHs—benzo(a)pyrene, BaA, BbF, and CHR (PAH4)—with a limit of less than 10 ppb as a more accurate food safety indicator.

EFSA was established precisely to separate scientific risk assessment from political decision-making and risk management, enhancing transparency, credibility, and scientific rigor in EU food safety governance. When超标 food is detected, the EU immediately initiates notification, traceability, sampling, recalls, and withdrawals. The EU’s 2023 regulation on maximum levels of specific contaminants in food explicitly prohibits the sale of contaminated food and bans its use as an ingredient or in mixing with other foods.

Chien questioned whether the 20% rule was established through a fair, impartial, and transparent process—one that all stakeholders, whether consumers, businesses, or officials, would willingly accept. Would we agree to a system where, after detecting benzo(a)pyrene above legal limits in cooking oil, a simple arithmetic calculation allows products with less than 20% tainted oil to remain on shelves?

He further questioned whether the government has the right to shift the responsibility for unsafe food from manufacturers to consumers without a comprehensive risk assessment. This exception was hastily decided in a short meeting involving experts, officials, and the contaminated oil manufacturer, raising serious concerns about procedural justice.

Chien concluded that the policy of not recalling food containing less than 20% benzo(a)pyrene-contaminated soybean oil does nothing to improve food safety and instead 'pours oil on the fire,' exacerbating the crisis. The real solution lies in establishing an independent, EU-style risk assessment system to align Taiwan’s food safety standards, risk evaluation, and management with international best practices.

He also warned that international food safety reports have detected high levels of PAHs in oils such as grape seed oil, olive pomace oil, unrefined vegetable oils, camellia oil, coconut oil, sesame oil, and sunflower oil. Since benzo(a)pyrene has been found in commonly used vegetable oils in Taiwan, it is crucial to investigate whether these oils could be significant dietary sources of benzo(a)pyrene, similar to soybean oil.

To prevent future food safety crises, Chien called for an independent root-cause investigation into the entire vegetable oil refining industry, including public and independent on-site factory inspections to identify the source of benzo(a)pyrene contamination.

FACT BOX

  • Source: PR Times
  • Category: Survey