Programs created to improve pregnancy outcomes have produced conflicting results. Most programs rely on screening questionnaires to identify patients at “high risk” for low-birth-weight infants or premature delivery. These high-risk patients are then enrolled in special programs, often involving clinic or home visits by nursing staff. Since the relationship between the nursing staff and the patients appears to be a crucial factor in determining pregnancy outcome, telephone counseling has been suggested as a cost-effective approach. One study found telephone support to be as effective as home uterine activity monitoring, but a second small study found no difference in pregnancy outcomes between the patients supported by telephone contact and the patients in the control group. Moore and colleagues conducted a randomized trial to assess the effect of regular telephone contact by nursing staff on preterm and low-birth-weight deliveries in women attending a public clinic.
All patients who spoke English and had access to a telephone were asked to participate in the study, beginning between 22 and 32 weeks' gestation. The 1,554 women who agreed to participate were randomly assigned to either the “phone” intervention group or the control group. Upon entering the study, all participants were given an extensive personal interview and received printed information on preventing preterm labor. Women in the phone group agreed to accept telephone contact three times per week, and arrangements were made to minimize the number of calls in which contact could not be established. Three trained nurses were responsible for placing all of the calls. The telephone calls did not follow a specified script but during each call, health status was assessed, appropriate health advice was given and patient concerns were addressed. Most of the calls focused on assessment of health status, including perceptions specific to pregnancy, and on more general topics, such as smoking, alcohol and drug use, and nutrition. The nurse gave recommendations specific to each patient based on the telephone assessment. Each call also allowed time for discussion of areas of concern to the patient. All data were collected by a nurse who did not know to which group the patients had been randomized. The primary outcome variable measured was the difference in the rates of premature and low-birth-weight infants.
The groups did not differ in any important variable (e.g., age, marital status, smoking, previous history of a low-birth-weight infant) at baseline. The median number of calls completed during the study ranged from 21 to 25, depending on the subgroup of patient when categorized by age and race. The median duration of each call ranged from 3.2 to 3.7 minutes. The rate of low birth weight (less than 2,500 g [5 lb, 8 oz]) was 10.9 percent in the intervention group, compared with 14.0 percent in the control group. Although this result did not reach statistical significance, the effect was most evident in black women 19 years of age or older. In these 759 women, the rate of low birth weight was 11.4 percent in the intervention group, compared with 17.3 percent in the control group. When analyzed by risk status, the reduction in low birth weight was most pronounced in women identified as low risk. For preterm births (less than 37 weeks' gestation), the overall rates were 9.7 in the intervention group and 11.0 in the control group but again, the effect was most pronounced in black women older than 19 years. In this group, the difference in preterm births (8.7 percent in the intervention group versus 15.4 percent in the control group) was statistically significant.
The authors conclude that although reductions in low-birth-weight and preterm births were achieved by the telephone intervention system, these reductions only reached statistical significance in black women 19 years of age or older. Much remains to be investigated on this topic. In particular, the optimal number and content of phone calls should be established, and the effect on large populations of women at relatively low risk should be studied.