This study presents a method to track changes in health service use in Kenya using imperfect national health management information system (HMIS) data. Researchers used a space-time geostatistical model to compensate for missing data from health facilities, allowing them to estimate monthly and annual outpatient service use. The analysis revealed a steady decline in health facility use from 1996 to 2002, followed by a sharp increase from 2003. This pattern was consistent across different causes of attendance and observed in all six major provinces. The method enables reliable estimation of national health service use patterns despite incomplete data, demonstrating the potential of HMIS data for monitoring health system performance. The approach was tested using data from 1,271 government-run and 402 faith-based health facilities between 1996 and 2004. The study highlights the importance of using geostatistical methods to improve the reliability of health data, which is crucial for evidence-based decision-making in resource-constrained settings. The findings suggest that changes in health policy, such as the introduction of the '10/20' fee policy in 2004, significantly impacted service use. The study also shows that the approach can be applied to other countries with imperfect national data to enhance monitoring of health service use and policy impact. The results demonstrate that HMIS data, when properly analyzed, can provide valuable insights into health system performance and inform policy decisions. The study underscores the need for improved data collection and analysis methods to support effective health system management.This study presents a method to track changes in health service use in Kenya using imperfect national health management information system (HMIS) data. Researchers used a space-time geostatistical model to compensate for missing data from health facilities, allowing them to estimate monthly and annual outpatient service use. The analysis revealed a steady decline in health facility use from 1996 to 2002, followed by a sharp increase from 2003. This pattern was consistent across different causes of attendance and observed in all six major provinces. The method enables reliable estimation of national health service use patterns despite incomplete data, demonstrating the potential of HMIS data for monitoring health system performance. The approach was tested using data from 1,271 government-run and 402 faith-based health facilities between 1996 and 2004. The study highlights the importance of using geostatistical methods to improve the reliability of health data, which is crucial for evidence-based decision-making in resource-constrained settings. The findings suggest that changes in health policy, such as the introduction of the '10/20' fee policy in 2004, significantly impacted service use. The study also shows that the approach can be applied to other countries with imperfect national data to enhance monitoring of health service use and policy impact. The results demonstrate that HMIS data, when properly analyzed, can provide valuable insights into health system performance and inform policy decisions. The study underscores the need for improved data collection and analysis methods to support effective health system management.