OBJECTIVE: To analyze patient complaints in a Primary Health Care District (PHCD) using statistical process control methods compared to multivariate methods, as regards their results and feasibility of application in this context. MATERIAL AND METHOD: Descriptive study based on an aggregate analysis of administrative complaints. SETTING: Complaints received between January 2005 and August 2008 in the Customer Management Department in the 3rd PHCD Management Office, Madrid Health Services. Complaints are registered through Itrack, a computer software tool used throughout the whole Community of Madrid. MAIN VARIABLES: Total number of complaints, complaints sorted by Reason and Primary Health Care Team (PHCT), total number of patient visits (including visits on demand, appointment visits and home visits) and visits by PHCT and per month and year. STATISTICAL ANALYSIS: Multivariate analysis and control charts were used. RESULTS: 44-month time series with a mean of 76 complaints per month, an increasing trend in the first three years and decreasing during summer months. Poisson regression detected an excess of complaints in 8 out of the 44 months in the series. The control chart detected the same 8 months plus two additional ones. CONCLUSIONS: Statistical process control can be useful for detecting an excess of complaints in a PHCD and enables comparisons to be made between different PHC teams. As it is a simple technique, it can be used for ongoing monitoring of customer perceived quality.
OBJECTIVE: To analyze patient complaints in a Primary Health Care District (PHCD) using statistical process control methods compared to multivariate methods, as regards their results and feasibility of application in this context. MATERIAL AND METHOD: Descriptive study based on an aggregate analysis of administrative complaints. SETTING: Complaints received between January 2005 and August 2008 in the Customer Management Department in the 3rd PHCD Management Office, Madrid Health Services. Complaints are registered through Itrack, a computer software tool used throughout the whole Community of Madrid. MAIN VARIABLES: Total number of complaints, complaints sorted by Reason and Primary Health Care Team (PHCT), total number of patient visits (including visits on demand, appointment visits and home visits) and visits by PHCT and per month and year. STATISTICAL ANALYSIS: Multivariate analysis and control charts were used. RESULTS: 44-month time series with a mean of 76 complaints per month, an increasing trend in the first three years and decreasing during summer months. Poisson regression detected an excess of complaints in 8 out of the 44 months in the series. The control chart detected the same 8 months plus two additional ones. CONCLUSIONS: Statistical process control can be useful for detecting an excess of complaints in a PHCD and enables comparisons to be made between different PHC teams. As it is a simple technique, it can be used for ongoing monitoring of customer perceived quality.