China’s State Administration for Market Regulation (SAMR, 市场监管总局) and three other national bodies said on 15 October 2026 that a joint investigation found no formamide in any baby diaper sample it tested. The investigation began on 22 June 2026 after online reports claimed that diapers contained the chemical.
All 168 domestic diaper producers were sampled and inspected
According to the joint investigation notice, the National Disease Control and Prevention Administration (国家疾控局) organised local disease control bodies to sample all 168 baby diaper producers in China, including contract manufacturers.
They collected 368 batches of finished diapers from 180 brands. A further 32 batches were tested by component, covering parts such as waistbands and fastening tabs on some leading brands. Formamide was not detected in either set of samples.
On-site inspections of the same 168 producers found that all met the hygiene code for manufacturers of disinfection products. Inspectors found no formamide added to, or contained in, production processes or raw materials.
| Item | Published detail |
|---|---|
| Issuing body | Joint investigation team of SAMR, the Ministry of Industry and Information Technology (工业和信息化部), the National Health Commission (国家卫生健康委) and the National Disease Control and Prevention Administration |
| Issue date | 15 October 2026 |
| Effective date | Not applicable; the notice reports investigation findings |
| Scope | Baby diapers made in China, third-party and self-media test claims, and blood tests at one Shandong centre |
| Investigation period | From 22 June 2026; the notice states the work has ended |
Retests of 58 batches held by a blogger found no formamide
The team sealed diapers left over from tests by a self-media blogger on Xiaohongshu (RedNote). Three laboratories, including the Chinese Center for Disease Control and Prevention, retested the samples whose packaging was intact: 58 batches from 15 brands. None contained formamide.
Experts organised by the Ministry of Industry and Information Technology reproduced the blogger’s test in full. They found the method had not been strictly validated and that the mass-to-charge data had not been assessed for deviation from formamide’s exact mass. The notice says no detection conclusion could be drawn.
Unit: batches
Batches tested in the investigation · Source: SAMR, 15 October 2026
View as table
| Item | Value |
|---|---|
| Finished products, 180 brands | 368 batches |
| Component tests | 32 batches |
| Blogger’s samples retested, 15 brands | 58 batches |
All 210 blood samples from a Shandong centre tested negative on retest
The notice says no method for detecting formamide in blood had been reported in China or abroad before the case. Blood testing at the Shandong Public Health Clinical Centre’s mass spectrometry unit (山东省公共卫生临床中心质谱研究与临床应用中心) began in June 2026 and was at a preliminary research stage.
Expert reviewers found that blank samples gave the same signal, signal-to-noise ratios fell short of basic qualitative requirements, data points were selectively chosen and raw data were deleted. They judged the results not credible.
Experts then developed a blood test method and retested all 210 sealed blood samples from the centre. None contained formamide. The Shandong Provincial Health Commission has dealt with violations by the medical institution and staff under relevant rules.
Third-party lab results showed a 7 percent spectral match
Hefei Guangce Product Testing Institute (合肥广测) lacked the capability to test for formamide and subcontracted the work to Suzhou Puze Testing Technology (苏州谱择). Expert review found no blank or parallel samples, no comparison of retention times with a reference standard, and only a 7 percent match between sample and formamide spectra.
Two report screenshots from Lianxin Testing (Jiangsu) (联信检测（江苏）) circulated online on 25 June, showing 414 mg/kg and 384 mg/kg. The notice says these were draft copies sent to the client for checking. Its reviewers found the peak’s retention time differed from formamide’s; the company withdrew the drafts and declared them invalid.
A second self-media account posted diaper review videos on Xiaohongshu, Weixin Channels and Douyin after 20 June. Its operator lacked testing capability and commissioned another person. Experts found the same types of problems, including no method validation and deleted raw data.
What the notice covers
- Sampling: 368 finished-product batches from 180 brands and 32 component batches, from all 168 domestic producers including contract manufacturers.
- Retests: 58 batches from 15 brands held by a blogger, and 210 blood samples from the Shandong centre.
- Laboratories named: the Chinese Center for Disease Control and Prevention, the China Academy of Quality Inspection and Testing Science and the Beijing Center for Disease Control and Prevention, among others.
- Follow-up: the departments say they will continue to strengthen supervision of infant products and guide producers to improve quality management systems.
