Same product, same sample, same batch. But take them to two different government labs for analysis, the results will vary by up to 766 percent or about nine times.

In late July, Bangladesh Food Safety Authority provided pasteurised milk samples of 14 top brands to three renowned local labs to determine the level of heavy metal and antibiotics in them.

The labs are Bangladesh Council for Scientific and Industrial Research (BCSIR), the Institute of Public Health (IPH) and Bangladesh Atomic Energy Commission (BAEC) -- all state-funded agencies.

But their findings have raised questions about their credibility.

Take the case of Pabna-based Ichhamoti Dairy and Food Products (PURA), one of the brands whose milk was tested.

The test at the BAEC found the lead level to be 0.208 microgram/kg as opposed to 0.024 mcg/kg at the BCSIR -- a 766 percent difference.

In Bangladesh, the permissible limit of lead in milk is 0.02 mcg/kg.   

Samples of state-owned Milk Vita was also tested. In the test, the BCSIR, popularly known as Science Lab, found that the milk contains 0.023 mcg/kg lead, which is 1.15 times the permissible limit. The BAEC found it to be 0.011 mcg/kg, or about half of what the Science Lab test found.

There is more. While the BCSIR test detected the level of Ciprofloxacin, an antibiotic used to treat infections in bone and joint, abdomen, respiratory tract and urinary tract, among other things, to be at 53.83 mcg/kg (which is still within the permissible limit of 100 mcg/kg) in all the 14 samples, the Institute of Public Health found no trace of it in any of the samples.

Based on the test results from the BCSIR, IPH and BAEC, the High Court recently banned the production and consumption of 14 milk brands. The Supreme Court subsequently stayed the HC order, essentially lifting the ban.

However, the wide gaps in the test results have raised serious questions about the authenticity of the results.

 “It is difficult to rely on the results of local laboratories because many of their machines are obsolete and they follow invalid methods. Reagents [substance for use in the analysis] are also adulterated. Hence the variation in results. The credibility of the country’s labs is at stake,” Mahfuzur Rahman, former chairman of BFSA, told The Daily Star.

‘SOMETHING IS WRONG’

The accuracy of any lab tests depends on three ‘Ms’ -- machine, method and the man behind the machine -- and none of these are often right in Bangladesh, experts say.

“If the same sample is tested in two different labs and they come up with totally opposite results, it means there is something wrong,” said Barun Kanti Saha, chief scientific officer and director at the BCSIR.

Latiful Bari, head of Food Analysis and Research Laboratory at Dhaka University’s Centre for Advanced Research in Sciences, agrees.

He said that the temperature during preservation and the preservatives used could also contribute to varying results.

“There might also be human errors, because the person behind the machine is very important,” he noted.

Also, government laboratories rarely calibrate their machines, another likely factor behind the massive differences, he said.

Calibration is a set of operations to ensure that the readings from the instrument are accurate and consistent with other measurements. It increases the reliability of the instrument, which is highly important in laboratory best practices.

The machine with which the BCSIR tested lead in the milk samples was installed in 2007-08. The IPH machines, both for testing lead as well as antibiotics, were set up between 2011 and 2013. The lead testing machine at the BAEC was installed in 2010-11, sources said.

All three labs conducted their last calibration late last year, officials said, requesting anonymity.   

The IPH and BAEC declined to comment on their findings.

NO ACCREDITATION

There are 54 registered labs in the country. But except for the BAEC, no other lab has certification from the Bangladesh Accreditation Board (BAB) for testing heavy metal in milk.

Sources in the BAB, formed in 2010, said only the BAEC has the accreditation to perform tests to determine the presence of heavy metals such as lead, chromium and cadmium in milk.

“Accreditation is voluntary. It cannot be imposed,” said Nasirul Islam, deputy director at the Board.

Accreditation confirms whether a particular lab has the capacity to run tests in line with international standards, he added. 

The Bangladesh Standard Testing Institution (BSTI) is the only designated government standard testing watchdog in the country. This means it should have all the means to test and confirm if a food item is safe.

Contacted, its Deputy Director for certification Reazul Haque said, “We have the accreditation to test 411 items. Heavy metal is not part of our safety parameters, so we do not run this test. Once we include heavy metals, antibiotics and pesticides in the parameters, we will bring in the necessary equipment to run these tests.”

But he added that having no accreditation does not necessarily mean that the test result is unreliable.

According to Mahfuzur, the former BFSA chairman, accreditation is essential because it is an assurance of quality control.

“In fact, it is not the labs that are accredited; the parameters are. Without accredited tests, we cannot say for sure that the test result is accurate,” he said.

Accreditation is important because it means the test is globally accepted, said Barun Kanti, the chief scientific officer at the BCSIR.