Describes a methodological study to evaluate the quality of care for an elderly population in the treatment of hypertension from a fairly simple analysis of medical records in primary care, with or without the use of additional information from patients, and shows how this type of data can be used to point out shortcomings in primary care. The data derive from a community survey of elderly people in one area of Jerusalem and relied on personal interviews, blood pressure measurements and an analysis of medical records over a period of four years. The measures used are: the percentage of persons without blood pressure (BP) measurement in a given year; the maximum “gap” between consecutive BP measurements in a given year; the rate of BP measurement per clinic visit; the percentage of hypertensives treated; and the percentage of hypertensives under control. The prevalence rates for hypertension in this elderly population vary between 40 per cent and 59 per cent according to the definition used. Using a simple and straightforward analysis of sick‐fund records several shortcomings in the surveillance and control of hypertension were detected. Calculation of the percentage of untreated hypertensives required a separate information source ‐ the screening. A BP survey is an important complement to an analysis of records. An ongoing evaluation of the quality of care of hypertensives in an elderly population should be based on screening, interviews with patients and analysis of the physician′s records. These provide an overall picture of the care in each medical practice enabling the physician to improve the quality of care provided to his or her patients.