Skip to main content

Risk Factors for Brucellosis during an Extended Outbreak in a Rural Hamlet, Medina, Saudi Arabia, 1994.


In October 1992, an outbreak of human brucellosis began in a small (2670 population) rural hamlet, Wadi Reem, in the Madina region. The principal economic activity of Wadi Reem was raising goats and sheep. Health and agricultural authorities advised residents to boil milk before drinking it and vaccinated their livestock. However, new brucellosis patients continued to appear at the local clinic and we started an investigation to learn the reasons for transmission.


We defined brucellosis as a febrile illness diagnosed by a physician as brucellosis and with a brucella agglutination titer ≥ 1:160, from October 1992 to September 1994 in a resident of Wadi Reem. We identified brucellosis patients from clinic and hospital records. We visited all 63 houses in the most heavily affected hamlet, Shalail, and interviewed families about exposure of all family members (446) to livestock.


Within the Shalail hamlet (attack rate 178 per 1000) there were 7 population clusters with attack rates ranging from 32 per 1000 to 476 per 1000. Age specific rates ranged from 580 per 1000 in females more than thirty years old to 46 per 1000 among children below 5 years old. We restricted the analysis to three clusters with an attack rate of 299 per 1000 and we stratified further analysis by age and sex. Brucellosis was related to livestock ownership (Risk Ratio [RR] = 1.4, 95% Confidence Interval [CI] 1.1-1.8). Among families that owned goats or sheep brucellosis was associated with entering livestock pens (RR = 1.5, 95% CI = 1.1-2.0) and the risk was higher among persons who reported skin lesions on their hands and feet (RR = 2.2, 95% CI = 1.7-3.0). Other livestock exposures such as being the family shepherd (RR = 1.0, 95% CI = 0.8-1.3), milking (RR = 1.3, 95% CI = 0.9-1.7), assisting with the birth (RR = 1.8, 95% CI = 0.8-1.4) or slaughtering (RR = 0.9, 95% CI = 0.6-1.5) were not associated with brucellosis. Brucellosis was not associated with drinking raw milk (RR = 1.1, 95% CI = 0.8-1.5).


The original control measures for brucellosis were probably incomplete. Contact with livestock crowded in pens and pericutaneous innoculation are probably more important than drinking raw milk. The focal nature of brucellosis in this community suggests that identification of infected herds will be an important aspect of brucellosis control.