Objective  To examine incidence trend and spatial distribution characteristics of pulmonary tuberculosis in 31 provinces, autonomous regions and municipalities directly under the central government in during 2012 – 2016, and to provide references for tuberculosis control in China.

  Methods  The data on pulmonary tuberculosis incidence in China were extracted from statistical yearbook for health and family planning in China from 2013 to 2017. Global and local spatial autocorrelation analysis were used to assess spatial correlation and spatial clustering characteristics of pulmonary tuberculosis incidence in China from 2012 to 2016.

  Results  The annual pulmonary tuberculosis incidence for the years from 2012 to 2016 were 70.62, 66.80, 65.63, 63.42, and 61.00 per 100 000 population in China, with a significantly downward trend (χ²_trend = 3.90, P = 0.048) and an average annual decline rate of 3.60%. The global spatial autocorrelation analysis resulted in the Moran′s I values for annual pulmonary tuberculosis incidence of 0.125, 0.129, 0.125, 0.112, and 0.105 for the years from 2012 to 2016 and there was a positive spatial correlation in annual pulmonary tuberculosis incidence (P < 0.05). The results of local spatial autocorrelation analysis revealed that the incidence of pulmonary tuberculosis mainly presented a high (low) – high (low) positive correlation clustering in eastern region, high (low) – low (high) inverse correlation clustering in central and western regions, and a higher regional spatial clustering in Beijing, Tianjin and Shandong province.

  Conclusion  The incidence of pulmonary tuberculosis showed a downward trend in 2012 to 2016 in China. There was a spatial correlation among regional incidences of pulmonary tuberculosis and there was an obviously regional clustering for pulmonary tuberculosis incidence in Beijing, Tianjin and Shandong province.