Carbon-block and jug filters can raise bacterial counts above the tap water going in, in some studies by a hundredfold or more. The WHO says this does not by itself indicate a health risk when the incoming water is safe, but calls the meaning for vulnerable people unclear. Changing cartridges on time is the practical answer.
| Source | What it found (quoted) and its limits | Evidence |
|---|---|---|
| Bartram J, Cotruvo J, Exner M, Fricker C, Glasmacher A (eds). Heterotrophic Plate Counts and Drinking-water Safety. WHO/IWA Publishing. (2003) · new Agency expert consensus review · International (WHO) · n = n/a |
'there is no direct relationship between HPC values in ingested water and human health effects in the population at large' (qualified 'in the absence of faecal contamination'); POU: 'Increases of HPC (due to growth) in these devices therefore do not indicate the existence of a health risk' 'as long as the entry water meets acceptable microbial water quality norms'; vulnerable people: implication 'is unclear' Limits: Consensus document, 2003; predates sequencing-based opportunistic pathogen studies |
Evidence: strong Free to read · source · added 2026-10-30 |
| Wu C-C, Love NG, Olson TM. Bacterial transmission and colonization in activated carbon block (ACB) point-of-use (PoU) filters. Environ Sci: Water Res Technol 7:1114-1124. (2021) · new Lab challenge study with 2 leading US carbon-block brands · Chloraminated tap water spiked ~5 x 10^2 cells/mL GFP-P. aeruginosa and E. coli; 29 days · n = 2 brands x duplicate filters |
P. aeruginosa in effluent 'within the first 6 L'; breakthrough at ~1.5% (P. aeruginosa) and ~7% (E. coli) of design-life volume; P. aeruginosa effluent plateau 'up to 1.8 times' influent; E. coli plateau ~35% of influent; authors: ACB filters 'not designed or certified to remove bacteria' Limits: Spiked challenge, not natural tap water; no HPC reported |
Evidence: moderate Access unknown · source · added 2026-10-30 |
| Wu C-C, Ghosh S, Martin KJ, Pinto AJ, Denef VJ, Olson TM, Love NG. The microbial colonization of activated carbon block point-of-use (PoU) filters with and without chlorinated phenol disinfection by-products. Environ Sci: Water Res Technol. (2017) · new Lab bench study with commercial faucet-mount carbon-block filters · Ann Arbor, MI municipal chloraminated tap water; 67 days; 16 h use/8 h stagnation · n = 3 pairs of filters |
Influent HPC mean '1.6 x 10^2 +/- 2.5 x 10^2 CFU mL-1'; effluent 'stabilized at an average of 2.3 x 10^4 +/- 1.0 x 10^4 CFU mL-1 after 40 days'; abstract: effluent 'plateaued at levels up to 100 times greater than influent counts'; Mycobacterium enriched Limits: Run to 137% of rated volume; one brand; 16S cannot confirm pathogenic strains |
Evidence: moderate Free to read · source · added 2026-10-30 |
| Chaidez C, Gerba CP. Comparison of the microbiologic quality of point-of-use (POU)-treated water and tap water. Int J Environ Health Res 14(4):253-260. (2004) · new Field sampling POU-treated vs tap water + QMRA · not stated in abstract · n = not stated in abstract |
HPC in all POU samples 'with concentrations ranging from 10 2 to 107 colony forming units/mL' (superscripts lost: 10^2-10^7 CFU/mL); P. aeruginosa, acid-fast organisms, total coliforms in '38.5, 43.8, and 82.4% of the samples'; colonization 'probability of less than 10-6' per day Limits: Abstract only; sample size not in abstract; risk model assumptions |
Evidence: moderate Paywalled · source · added 2026-10-30 |
| Daschner FD, Ruden H, Simon R, et al. Microbiological contamination of drinking water in a commercial household water filter system. Eur J Clin Microbiol Infect Dis 15:233-237. (1996) · new Field survey + lab test of real commercial household filters · Germany; household tap water · n = 34 household filters; 6 lab filters; 13 new filters |
'In 24 of 34 filters used in households, bacterial counts increased in the filtered water up to 6,000 cfu/ml'; 'In some cases colony counts in the filtered water were 10,000 times those in tap water'; 'filter material of 5 of 13 new commercial filters was contaminated with bacteria or moulds' Limits: Abstract only; 1990s product generation; HPC not a direct health endpoint |
Evidence: moderate Paywalled · source · added 2026-10-30 |
| Snyder JW Jr, Mains CN, Anderson RE, Bissonnette GK. Effect of point-of-use, activated carbon filters on the bacteriological quality of rural groundwater supplies. Appl Environ Microbiol. (1995) · new Field study of installed POU powdered-activated-carbon filters · Rural domestic spring and well supplies (location not in abstract) · n = 24 domestic supplies |
After overnight stagnation HPC in PAC-treated effluents 'elevated by 0.86 and 0.20 orders of magnitude for spring and well water systems'; after flushing 'for 2 min' HPC 'significantly reduced (P < 0.01) below influent levels' Limits: Abstract only; untreated groundwater, not chlorinated municipal water |
Evidence: moderate Access unknown · source · added 2026-10-30 |
| Chen Y, Fu J, Wang Y, et al. Microbial risks in household UF purifiers and disinfection strategy with electrolysis. E3S Web of Conferences 143:02028. (2020) · new Lab stagnation test of household ultrafiltration purifiers · China · n = not stated |
'The level of HPC exceeded the hygienic standard when the stagnation time was more than 3 h'; 'lag phase of bacterial growth was no more than 12 h in filters' Limits: Conference paper; abstract only; UF not RO |
Evidence: weak Free to read · source · added 2026-10-30 |
| Yi J, Lee J, Jung H, Park P-K, Noh SH. Reduction of bacterial regrowth in treated water by minimizing water stagnation in the filtrate line of a gravity-driven membrane system. Environ Eng Res 24(1):17-23. (2019) · new Lab study of household gravity-driven membrane unit · Korea · n = not stated |
HPC 'finally increased up to 10^5CFU/mL'; HPC of filtrate 'rapidly decreased within 5 min for each batch filtration'; 'the filtrate produced at the first 5 min is recommended not to be used as potable water' Limits: Abstract only; prototype system, not a commercial US product |
Evidence: weak Free to read · source · added 2026-10-30 |
| Authors not retrieved (Tabriz University of Medical Sciences repository). Bacteriological quality of water produced by household water treatment devices. (2015) · new Cross-sectional inlet/outlet sampling of household devices · Tabriz, Iran (2014) · n = 18 devices; 36 samples |
'Total and fecal coliform bacteria were negative at all inlet samples, but were positive at two of the outlet samples'; 'HPC were higher than the maximum limits of 500 cfu/ml at 67% of the outlet samples' Limits: Abstract only; journal/authors not identified; small n; device types unspecified |
Evidence: weak Access unknown · source · added 2026-10-30 |
Evidence grade: strong = trials, large field studies or official datasets; moderate = smaller or lab studies with real water; weak = single small studies, makers' specs or secondary reports. "Free to read" means the source itself is free to read online.