Results

Sixty two women were recruited for the UC group, and 52 women for the MI group. Participant characteristics are listed in table 1. There was a trend for participants in the UC group to be younger than those in the MI group, and significantly fewer of the UC participants were married. There was also a greater number of recent quitters in the UC intervention; there were no other differences between study groups.

The study outcomes are presented in table 2. Analyses were adjusted for baseline differences between groups, and intention to treat analyses were used. There were no significant differences in cessation rates between the two groups. In addition, no differences were found in follow up smoking rate. Although overall attrition rates were reasonable given the socioeconomic status and mobility of this population (35% by postpartum), there was a higher attrition rate in the intervention group (48% in interventionv 30% in control), which further reduced our power to evaluate study outcomes.

Discussion

The motivational intervention was not found to increase smoking cessation rates beyond the levels found in the usual care group. Although we adjusted for baseline differences in the analyses, it is important to note that there were differences in key demographic characteristics related to smoking between the two groups. Significantly more participants in the UC group were recent quitters, which is a predictor of postnatal smoking status. In addition, a larger percentage of women in the MI group were nicotine dependent; although this difference did not reach significance, it is worthy of note and may have made it more difficult for women in the MI group to consider quitting.

There were a number of important lessons learned in this pilot study. First, recruitment of very low income smokers to a smoking intervention was very challenging. Because of the structure of the intervention, we had to rely on HBP staff to recruit their clients to the study. Although HBP staff received a substantial amount of training on recruitment and on refocusing recruitment efforts away from smokingcessation, the study was often represented as a smoking cessation intervention, which likely affected participation rates. Second, the intervention dose delivered was less than planned. The HBP nurses reported that they face a number of challenges in working with their clients, including difficulty contacting them because of lack of working phones, and difficulty tracking them owing to frequent and often unexpected moves. Resources were not available to increase the number of contact attempts in order to deliver the smoking intervention, and thus the intervention dose delivered was less than optimal. Third, the time available to train the Healthy Baby Program nurses in the motivational intervention was limited. Nurses were initially trained in motivational interviewing in an 8 hour training session, with three, 2 hour follow up training sessions. In addition, follow up consultation was provided on an individual basis. However, during the individual follow up consultation it was clear that more intensive training would have been helpful, and that many program staff found it difficult to utilise a motivational style. One solution to this challenge would be to select nurses within a program who have good counselling skills, particularly related to motivational enhancement, and an interest in smoking, and select these nurses as “champions” who would deliver the intervention to all eligible pregnant women.

It is often challenging for outside research groups to gain trust with community based service programs and the community itself. It is essential that researchers develop their relationships with the community over an extended period of time; however, the requirements of a two year research project such as this greatly limited the amount of relationship development that could occur outside of the research needs, and thus may have impacted on the acceptance and delivery of the intervention. This is a challenge that researchers and funders alike need to address.

Overall, patients reported being quite satisfied with the intervention. In addition, the providers enjoyed having the opportunity to learn new skills, and found the motivational interviewing approach to be useful for working with clients on smoking as well as other behaviours. Although it is disappointing that the motivational intervention did not impact on smoking cessation rates among study participants, the findings of this pilot study suggest that perhaps more intensive interventions, with more extensive training of delivery staff, are needed to target smoking among very low income populations. The results of this study should not be interpreted as an indictment of a motivational intervention model, but rather as a call to conduct more extensive research into the underlying contextual factors that may influence smoking among low income pregnant women, and an effort to develop new intervention models that target these factors.