Identification of pregnancies that are at greater than average risk is a fundamental component of antenatal care. The objective of this study was to assess the level of appropriate management and outcomes among mothers with hypertensive disorders of pregnancy, postdates and anemia in pregnancy, and to determine whether the colour coding system had any effect on the maternal mortality ratios. A retrospective follow-through study confined to users of government health services in Peninsular Malaysia was carried out in 1997. The study areas were stratified according to their high or low maternal mortality ratios. The study randomly sampled 1112 mothers out of 8388 mothers with the three common obstetric problems in the selected study districts. The study showed that the prevalence of anemia, hypertensive disorders in pregnancy and postmaturity among mothers with these conditions were according to known international standards. There was no significant difference in the colour coding practices between the high and low maternal mortality areas. Inappropriate referrals were surprisingly lower in the areas with high maternal mortality. Inappropriate care by diagnosis and by assigned colour code were significantly higher in the areas with high maternal mortality. The assigned colour code was accurate in only 56.1% of cases in the low maternal mortality areas and in 55.8% of the cases in the high maternal mortality areas and these two areas did not differ significantly in their accurate assignment of the colour codes. The colour coding system, as it exists now should be reviewed. Instead, a substantially revised system that takes cognisance of evidence in the scientific literature should be used to devise a more effective system that can be used by health care personnel involved in antenatal care to ensure appropriate level of care and referrals.