BACKGROUND: High nasopharyngeal pneumococcal carriage density is associated with severe pneumonia; however, little is known about factors that affect pneumococcal carriage density including pneumococcal vaccination. We describe pneumococcal density by clinical and demographic factors, and effect of 13-valent pneumococcal conjugate vaccine (PCV13) on density in Papua New Guinea (PNG), Lao People’s Democratic Republic (Lao PDR) and Mongolia, 3–6 years following national PCV13 introduction. METHODS: Three prospective pneumococcal carriage surveillance studies enrolled children aged 2–59 months with acute respiratory infections in Lao PDR (2013–2019), and pneumonia in PNG (2016–2019) and Mongolia (2015–2019). Demographic and clinical factors were collected on interview and from medical records. Nasopharyngeal swabs were tested for pneumococci using lytA real-time quantitative PCR and molecular serotyping using DNA microarray. In unvaccinated children median pneumococcal carriage density was compared across relevant clinical and demographic factors using Wilcoxon rank sum or Kruskal Wallis tests. Quantile regression models were used to determine the association between pneumococcal density, vaccination status and number of PCV doses. RESULTS: A total of 1009 (PNG), 532 (Lao PDR) and 621 (Mongolia) pneumococcal carriers were included. Of carriers with serotyping results, PCV13 serotype (VT) carriage was 36.1% (356/985) in PNG, 40.8% (189/463) in Lao PDR and 50.7% (270/532) in Mongolia. The median pneumococcal VT density was 6.25 log10GE/ml (genome equivalents per milliliter) (interquartile range [IQR] 5.66, 6.79) in PNG, 5.74 log10GE/ml (IQR 4.99, 6.40) in Lao PDR and 5.64 log10GE/ml (IQR 5.11, 6.32) in Mongolia. In PNG, Lao PDR and Mongolia, 54.4%, 51.1% and 34.9% pneumococcal carriers were fully vaccinated, respectively. There was no difference in VT pneumococcal density by relevant clinical and demographic factors in unvaccinated children. In PNG, VT density was slightly lower (-0.36, 95% confidence interval [CI] -0.61, -0.12; p = 0.004) among vaccinated compared with unvaccinated children, in particular those who received three doses (-0.37 95% CI -0.63, -0.10; p = 0.007). No differences were observed in Lao PDR and Mongolia. CONCLUSIONS: We demonstrated variable results across our three sites. Indirect PCV13 effects may have resulted in limited observed reductions in VT density in unvaccinated children. In PNG, PCV13 vaccination was associated with a decline in VT density. TRIAL REGISTRATION: Not applicable. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12879-025-12328-w.