To the Editor: The population of the world that is more than 65 years of age is rising by 2.5% per annum. Most of this growth is projected to occur in so-called developing countries.1 However, current health policies focus on infant and maternal health. There have been few published studies of older people in Africa, and their health problems are ill-defined.2 We screened the health of the older residents of three villages, Keneba, Kanton Kunda, and Manduar, The Gambia, West Africa. Keneba has been the setting for tropical health and nutrition research by the UK Medical Research Council (MRC) for more than 40 years.3 The villagers are Muslim rural subsistence farmers. They live in single-story houses within compounds based on family ties. There is no electricity or running water. All of the villages' citizens were identified by a census. A representative sample of 106 subjects (55 women and 51 men) older than 59 years of age was selected using random number tables. Subjects were interviewed at home by a field-worker using a structured questionnaire. Data were collected on socioeconomic status and personal health. Subjects were asked about their experience of 18 symptoms over the preceding 14 days.4 An assessment of daily living (ADL) score was derived from a standard questionnaire.4 Locally appropriate questions about physical independence — bathing, dressing, feeding, walking outdoors — were asked, as were questions relating to instrumental independence — fetching water, making a fire, and working in the fields. A nine-question abbreviated mental test score (MTS) was also devised.4 A clinical examination was performed in the subjects' own homes. Results are presented as summary statistics for men and women. Differences between the sexes were analyzed using X2, Fisher's exact and Mann Whitney tests where appropriate. The total population of the three villages was 3054, of whom 2298 were resident during the study. 7.2% of the population were aged greater than 59 (124 women and 98 men), comprising 9.7% of the residents. Data from the questionnaire are summarized in Table 1. More than 90% of subjects were married. Divorce was more common among men (P < .01); widowhood was more common among women (P < .001). Two-thirds of subjects were working. The most commonly reported symptoms were tiredness (84%), headache (71%), fever (66%), poor memory (63%), poor appetite (62%), and visual problems (59%). As noted by Heikkinen,4 women reported more symptoms than men (median of 8 versus 6 (P < .05)). Headache (P < .001) and poor appetite (P < .002) were more common in women. Urine problems (P < .001) and loss of power (P < .05) were more common in men. However, among Gambian men, this was probably misinterpreted as impotence. The distribution of ADL scores was skewed. The maximum of 21 was scored by 43%, and 74% scored higher than 18 of 21. Fifty-four percent of subjects had the maximum MTS score. The median for men was higher than for women (P < .005), which probably reflects a difference in formal education. There was no significant relationship between reported symptoms or clinical findings and MTS or ADL scores. The results of physical examination are shown in Table 2. The prevalence (95% confidence interval) of severe dental caries was 79% (71/87); 72% (68/76) had cataract; 59% (54/64) had corneal scarring; 29% (20/38) were hypertensive (BP greater than 159/94 mm Hg). Signs of osteoarthritis were present in 71% (63/80) and more common in women at the shoulders (P < .002) and the knees (P < .005). That the older people in this community remain independent is confirmed by high ADL scores. There is a high level of social and economic support from the extended family, similar to that found in Zimbabwe by Wilson.2 For more than 20 years, European doctors have provided free treatment at the MRC clinic, with proven benefits to children and women of child-bearing age.5 However, though 80% of our elderly subjects had been seen within the previous 6 months at the clinic, 42 (40%) had untreated conditions likely to benefit from treatment. The combination of examining the eyes and blood pressure and urine testing identify the majority of treatable conditions, and this level of screening does not require medical staff. It could be performed by community nurses working in rural clinics. Unfortunately, ADL, MTS, and symptom scores are not sensitive at the individual level in screening for those with medical needs. As the older African population grows, the need for treatment of noncommunicable diseases will increase. Appropriate training of medical and paramedical staff is required, and the costs of treatment must be anticipated.