UTILIZING AN INTERVIEW, A QUESTIONNAIRE, AND OBSERVATIONS DURING LABOR, DELIVERY, AND THE POSTPARTUM PERIOD, A SAMPLE OF 100 MOTHERS WAS IDENTIFIED AS AT HIGH RISK FOR ABNORMAL PARENTING PRACTICES. THESE MOTHERS WERE RANDOMLY DIVIDED INTO A 'HIGH-RISK INTERVENE' AND A 'HIGH-RISK NONINTERVENE' GROUP. THE INTERVENE GROUP RECEIVED COMPREHENSIVE PEDIATRIC FOLLOWUP BY A SINGLE PHYSICIAN, A LAY HEALTH VISITOR, AND A PUBLIC HEALTH NURSE IN THE HOME. THE NONINTERVENE GROUP RECEIVED ROUTINE CARE, ALTHOUGH THE RESULTS OF THESE SCREENING PROCEDURES WERE SHARED WITH THE PHYSICIANS AND NURSES RESPONSIBLE FOR THEIR CONTINUING CARE. ANOTHER GROUP OF 50 MOTHERS WHO DELIVERED DURING THE SAME TIME PERIOD AND WHO WERE ASSESSED AT LOW RISK IN TERMS OF ABNORMAL PARENTING PRACTICES SERVED AS CONTROLS. WHEN THE CHILDREN WERE APPROXIMATELY 2 YEARS OLD, 25 FAMILIES IN EACH OF THE 3 GROUPS WERE CHOSEN AT RANDOM FOR DETAILED EVALUATION. IT WAS FOUND THAT THE PERINATAL SCREENING PROCEDURES WERE SUCCESSFUL IN IDENTIFYING THE HIGH-RISK GROUP; THESE CHILDREN HAD ENCOUNTERED SIGNIFICANTLY DIFFERENT PARENTING PRACTICES THAN HAD THE CONTROL GROUP. FIVE CHILDREN IN THE NONINTERVENE GROUP REQUIRED HOSPITALIZATION FOR SERIOUS INJURIES THOUGHT TO BE SECONDARY TO ABNORMAL PARENTING PRACTICES, CONTRASTED WITH NO SUCH HOSPITALIZATIONS IN EITHER THE INTERVENE OR CONTROL GROUPS. LABOR-DELIVERY OBSERVATIONS AND NURSERY INTERVIEWS AND OBSERVATIONS PROVIDED THE MOST ACCURATE PREDICTIVE INFORMATION, WHILE PRENATAL INTERVIEWS AND QUESTIONNAIRES DID NOT ADD SIGNIFICANTLY. PERINATAL ASSESSMENT AND SIMPLE INTERVENTION WITHIN FAMILIES AT HIGH RISK SIGNIFICANTLY IMPROVED THE INFANTS' CHANCES FOR ESCAPING INJURY. ASPECTS OF THE PROJECT'S METHODOLOGY ARE DETAILED. WARNING SIGNS MANIFEST DURING THE PRENATAL, DELIVERY, AND POSTPARTUM PERIODS ARE PROVIDED. CHARACTERISTICS OF POSITIVE FAMILY CIRCUMSTANCES AND INTERVENTION TECHNIQUES ARE INCLUDED. A TABULAR SUMMARY OF THE STATISTICAL ANALYSIS AND SELECTED REFERENCES ARE PROVIDED. (KBL)