There is a growing body of evidence suggesting the importance of the management of maternal depression in primary care practices [15
]. In order to inform future intervention development, the present study examined potential determinants of management of maternal depression practices in primary care settings. As previously published, nearly 40% of the PCPs in this sample reported never or rarely assessing for maternal depression, less than 30% reported current use of a screening tool for maternal depression, and approximately 60% reported rarely or never providing counseling or referring patients who are depressed for follow-up mental health care [20
]. Clearly there is a need to better understand the determinants of maternal depression management practices to inform future intervening efforts.
Based on our conceptual model (see ), we proposed that PCPs' beliefs, attitudes, self-efficacy, and knowledge would predict management of maternal depression and that particular barriers may impede the likelihood of this behavior as well. Knowledge via past training (i.e., basic training or CME) for managing maternal depression seems to play an integral role in whether a physician will engage in maternal depression management practices. Our findings suggest that physicians who reported better training and higher levels of knowledge were more likely to report actively managing maternal depression in their practices. Unfortunately, over 60% of the PCPs reported that their past training in diagnosis or treatment of maternal depression was either fair, poor, or never received [20
]. These findings, coupled with past research suggesting that mental health training is limited within residency training, and that physicians often report that they do not feel they have had adequate training to correctly manage depression highlight the need for more education and training of primary care practitioners in the area of maternal depression [22
In addition to knowledge as a predictor, our findings suggest that self-efficacy and attitudes mediate the relationship between knowledge/past training and management practices. Self-efficacy, assessed in this study by level of perceived confidence and comfort in managing maternal depression, is often cited as a strong behavioral change agent closely tied to health behavior outcomes [30
]. Our study also found PCPs' perceived self-efficacy and attitude (i.e., feeling responsible for identifying maternal depression) as strong predictors of maternal depression management. Conversely, a lack of perceived confidence and a sense of feeling responsible to manage depression have been cited as barriers to managing depression in primary care practices [23
]. Not surprisingly, self-efficacy and attitudes related to feeling responsible for identifying women at risk for depression were strongly influenced by knowledge in the form of basic training and to some extent by specialized continuing education.
Attitudes related to the positive perception of mental health services were influenced by knowledge/past training and also linked to self-efficacy such that physicians who reported higher self-efficacy also perceived available mental health services more favorably. This is of particular relevance since past research has reported provider dissatisfaction with current mental health management practices as a systemic barrier toward managing depression in primary care settings [22
]. This reinforces the importance of increasing knowledge via basic training and CEU attainment focusing on maternal depression. Experts in implementation research suggest that this training should extend beyond standard education and skill-based training and include ongoing consultation and coaching [32
]. Providing ongoing consultation and coaching is one of six integral components (i.e., staff selection, preservice and inservice training, staff and program evaluation, facilitative administrative support, and systems interventions) in implementing successful change within organizations [32
]; in this case, change in PCPs depression management in primary care practices.
Contrary to what we hypothesized, physician beliefs pertaining to perceived impact of maternal depression were not a strong predictor in the model. This outcome is most likely due to the lack of variance in response, as more than 95% of the respondents believed maternal depression to be detrimental. However, the belief that maternal depression goes away without treatment was negatively correlated with management practices.
Surprisingly, the present study did not find often cited barriers related to limited time and financial issues to be determinants of maternal depression management practices. In terms of financial barriers, only 10% of the sample endorsed this item whereas it was more of an even split related to time. For both factors, measurement most likely impacted our findings as both potential factors were measured by dichotomous (yes/no) items instead of a Likert-type measurement. The other possibility may be that other elements in the model, like perceived responsibility, washed out any influence that time or financial considerations might have had. Financial issues have been suggested to play a role as capitation rates often exclude mental health services and few incentives exist for mental health specialists to collaborate with primary care providers [33
]. However, contrary to what we expected, barriers related to financial issues were not strong predictors of maternal depression management practices.
Findings from our study should be evaluated in light of the following limitations. Despite using techniques linked to optimal response rates [34
], our response rate was suboptimal (see the work of Leiferman et al. (2008) [20
] for more on sampling and nonresponse limitations). Findings from this study have limited generalizability as they are based on a small, geographical sample and thus, may not represent other populations. Moreover, this was the first model to be tested in this population, thus may further limit the generalizability of the model; although only relatively strong effects were reported here. Given the small sample size we were not able to conduct models examining differences across the three specialties (i.e., obstetrics, pediatrics and family medicine), thus, more research is warranted in this area.
Modeling efforts from this study suggest that much of PCPs' practices in identifying and treating maternal depression, when taken together, are tightly linked to attitudes related to feelings of responsibility for those patient issues and self-efficacy/confidence with maternal depression management. Basic training and continuing education in maternal depression seem to be integral in this process as they have a direct impact on management practices, but also indirectly affect management practices through perceived self-efficacy and responsibility.
Overall PCPs reported that they are open to making modifications to their practice and improving their knowledge and skills related to managing maternal depression. [20
] Implementing screening protocols is one potentially effective way to identify maternal depression and initiate prompt treatment of the disorder. A recent study found screening for maternal depression during well child visits to be feasible [35
]. Improved screening coupled with strong coordination with mental health services would represent a significant advance in reducing maternal and child health risks and is likely to improve health outcomes. In addition to implementing screening tools, innovative and novel models need to be developed and revised to promote the management of maternal depression. The findings from our previous study suggest that PCPs in general would like more training on mental health topics in the form of CEUs, guidelines, and computer deliverables [20
]. In particular, the majority of the PCPs stated they would like information to enhance patient communication about mental health issues. In conclusion, the present study's findings may ultimately help inform the design of future intervention models aimed at improving the management of maternal depression in primary care practices.