An Integrated Web-Based Mental Health Intervention of Assessment-Referral-Care to Reduce Stress, Anxiety, and Depression in Hospitalized Pregnant Women With Medically High-Risk Pregnancies: A Feasibility Study Protocol of Hospital-Based Implementation

Background At prevalence rates of up to 40%, rates of depression and anxiety among women with medically complex pregnancies are 3 times greater than those in community-based samples of pregnant women. However, mental health care is not a component of routine hospital-based antenatal care for medically high-risk pregnant women. Objective The purpose of this study is to evaluate the effectiveness and feasibility of the hospital-based implementation of a Web-based integrated mental health intervention comprising psychosocial assessment, referral, and cognitive behavioral therapy (CBT) for antenatal inpatients. Methods This study is a quasi-experimental design. Pregnant women are eligible to participate if they are (1) <37 weeks gestation, (2) admitted to the antenatal inpatient unit for >72 hours, (3) able to speak and read English or be willing to use a translation service to assist with completion of the questionnaires and intervention, (4) able to complete follow-up email questionnaires, (5) >16 years of age, and (6) not actively suicidal. Women admitted to the unit for induction (eg, <72-hour length of stay) are excluded. A minimum sample of 54 women will be recruited from the antenatal high-risk unit of a large, urban tertiary care hospital. All women will complete a Web-based psychosocial assessment and 6 Web-based CBT modules. Results of the psychosocial assessment will be used by a Web-based clinical decision support system to generate a clinical risk score and clinician prompts to provide recommendations for the best treatment and referral options. The primary outcome is self-reported prenatal depression, anxiety, and stress symptoms at 6-8 weeks postrecruitment. Secondary outcomes are postpartum depression, anxiety, and stress symptoms; self-efficacy; mastery; self-esteem; sleep; relationship quality; coping; resilience; Apgar score; gestational age; birth weight; maternal-infant attachment; infant behavior and development; parenting stress/competence at 3-months postpartum; and intervention cost-effectiveness, efficiency, feasibility, and acceptability. All women will complete email questionnaires at 6-8 weeks postrecruitment and 3-months postpartum. Qualitative interviews with 10-15 health care providers and 15-30 women will provide data on feasibility and acceptability of the intervention. Results The study was funded in September, 2014 and ethics was approved in November, 2014. Subject recruitment will begin January, 2015 and results are expected in December, 2015. Results of this study will determine (1) the effectiveness of an integrated Web-based prenatal mental health intervention on maternal and infant outcomes and (2) the feasibility of implementation of the intervention on a high-risk antenatal unit. Conclusions This study will provide evidence and guidance regarding the implementation of a Web-based mental health program into routine hospital-based care for women with medically high-risk pregnancies.


January 24, 2013
Dr. Dawn Aileen KINGSTON 5-258 Edmonton Clinic Health Academy University of Alberta 11405-87th Ave Edmonton, Alberta T6G 1C9 Dear Dr. KINGSTON: We are pleased to inform you that the Canadian Institutes of Health Research (CIHR) has approved your recent application entitled "Integrated Maternal Psychosocial Assessment to Care Trial (IMPACT): Intervening Early to Improve Maternal and Child Health". If you are receiving this letter through ResearchNet, your Authorization for Funding will follow in the mail otherwise it is enclosed in this package.
If you have not already received the review documents related to your proposal, please contact us. Should you have any questions about the review process, please address them directly to CIHR staff. Do not contact the officers or members of the peer review committee. As CIHR does not notify co-applicants of the decision, we ask that you inform those individuals involved, along with their research institutions (if different from your own), of the outcome of this application.
Congratulations on your success in this competition. You should take great pride in your success, particularly in light of the very competitive nature of CIHR peer review.
As you know, peer review is the cornerstone of our research funding system. This process rests on the kind voluntarism of your colleagues at other institutions who generously gave their time to review your application.
The Canadian Institutes of Health Research is committed to building an innovative national health research enterprise. To this end we have initiated a process to design a new Open Suite of Programs and peer review system to ensure the long-term sustainability of CIHR's contribution to the Canadian health research enterprise, to remove barriers, and to enable researchers from all pillars to improve CIHR's ability to deliver on its mandate.
To meet CIHR goals, we must also continue to communicate the value of health research to Canadians. That is why we encourage you to work with your institution to promote your research . We have developed guidelines on public communication, available on our website at http://www.cihr-irsc.gc.ca/e/30789.html, to support you in this activity.
Once again, congratulations and I wish you success in your research. o purpose of the proposal: to evaluate the acceptability, clinical-and cost-effectiveness of a sustainable, innovative, integrated process of online psychosocial for pregnant women -focused on depression and anxiety. Intervention is an online CBT program assessment, referral, and cognitive behaviour therapy (CBT) comprising 6, 30-minute modules in 272 pregnant women prior to 32 weeks gestation at non-high risk for depression, anxiety, stress o hypothesis to be tested, or the research questions to be answered: early interventioncan improve maternal mental health in pregnancy/postpartum and should improve stress outcomes o objectives to be achieved and approach proposed: in a randomized controlled trial and interviews, main outcome is reduced symptoms of prenatal depression/anxiety/stress (minimal clinically important difference of 4 points in each of the depression, anxiety, and stress subscales of the DASS21). Secondary clinical outcomes are reduced risk of poor child (birth weight, Apgar scores), maternal (postpartum depression/anxiety/stress, parenting competence, parenting stress, maternal-child attachment, sleep quality) and family outcomes (partner relationship, maternal-child attachment). There is a prospective economic evaluation, i.e. a within-trial cost effectiveness analysis comparing the integrated intervention 'package' with usual prenatal care. The perspective of the primary economic evaluation will be that of the health and social care budget; a secondary analysis will adopt a societal perspective incorporating personal costs and productivity costs in addition to the health and social costs associated with the delivery of the intervention and subsequent service utilization by study participants. Project has 2 phases: (1) evaluation of the integrated psychosocial assessment-referral-CBT intervention (RCT); and (2) assessment of the feasibility, acceptability, and mechanisms of the intervention (qualitative interviews). o progress made to date: findings of a pilot for his IMPACT-trial revealed that women found the program acceptable with recommendations that it be offered: (1) in pregnancy; (2) in an online format; and (3)  · This is an evaluation of a complex intervention, and the intervention is unblended, and outcome is self reported… Good address of potential sources of bias, though, need to prevent spill-over of information and expectations with women in the control group

Synopsis
The overall research question in this grant is "Does a prenatal cognitive behavioral care program reduce maternal stress and improve maternal and infant outcomes?" The applicants stress this project is about assessing integrated care of assessment-referral-treatment. In brief, assessment refers to using a screening tool to assess risk. High-risk mothers are referred to family physician, low-moderate risk mothers are referred to cognitive behavioral therapy (CBT) and no-risk mothers are provided their results and followed up but no treatment provided.
The intervention "group CBT" is reported effective in pregnant and new mothers. However, resources are limited. The applicants argue that mothers with low-moderate symptoms are unable to access CBT and are a vulnerable population and need to be able to access CBT on their own. Therefore, this grant proposes to develop an internet-based version of a successful CBT program that is delivered in 6 30-min modules over 4-6 weeks.

Strengths and Weaknesses
The applicants are using an RCT design to evaluate an intervention on an important clinical question, and supplementing the quantitative analysis with qualitative interviews to determine acceptability of the intervention. The internet-based intervention has the potential to reach a much larger population than is achieved with current methods. One weakness of this study is that the vulnerable population is being defined based on scores from a screening tool. These tools were specifically developed for screening -cut offs for scores should not be considered diagnostic. Therefore, I would prefer more evidence than what is currently in the grant about how "vulnerable" the proposed vulnerable population truly is. The analysis section could be clearer. The subgroup analyses appear to include post-randomization variables in the regression model. This would be appropriate for decomposing total causal effects into direct and indirect, but it appears the applicants are planning to interpret the statistical model as a total causal effect.

Research Question
Does a prenatal cognitive behavioral care program reduce maternal stress and improve maternal and infant outcomes? More specifically, the applicants want to determine clinical effectiveness of their internet-based CBT program, evaluate if integrated psychosocial care is efficient and feasible and cost-effective, and determine mediators and moderators of the intervention effect.

Literature Review
The applicants suggest that depression, anxiety and stress are under detected, and many women never seek treatment. The applicants argue that because symptoms begin during pregnancy, effective treatment is necessary at that time but do not provide any evidence of previous treatments.

Research Design
This is an RCT that includes qualitative interviews at a later date to determine feasibility, acceptability and mechanisms.

Research Methods
The inclusion criteria are pregnant women between 12-22 weeks gestation. Women with no stress are included because they could develop stress later. Women at high risk will be analyzed but treatment is referral to family physician (4-10% of subjects).
Mothers are recruited from one maternity clinic with 5 family physicians. First eligibility survey completed online with tablet in office. The applicants propose to use psychosocial assessment using screening tools (Edinburgh Postnatal Depression Scale (EPDS) and Antenatal Risk Questionnaire (ANRQ)). One limitation is that these are screening tools. Although many publications interpret them as clinical measures of stress and depression, they are not.
The CBT program itself had been pilot tested in paper form and adapted based on feedback in a pilot study. However, this intervention is an online version of the program, involving 6 30-min modules to be completed over 4-6 weeks. The online version is not yet developed. The control group gets usual care.
Outcomes of stress will be measured at 6-8 weeks post-randomization and 12 weeks postpartum: maternal stress, self-efficacy, self-esteem, sleep, and coping skills. Secondary outcomes include presence of (% above cutoff) and severity (mean score) for postpartum depression, anxiety, stress, mean prenatal and postnatal self-efficacy, sense of mastery, self-esteem, sleep, relationship quality, coping, APGAR, birth weight, maternal-infant attachment, infant behavior, parenting stress and cost effectiveness.
The sample size for the RCT appears adequate, but the structural equation modeling appears insufficient. The rule of thumb as stated is not quite accurate: it is that one should have at least 10 cases per level of a variable, not per variable (a variable with 3 levels requires 20 participants). Further, the definition of case for a dichotomous outcome such as the secondary outcomes is the number of positive cases, not the total population. The explanation is simple: if you have 50 cases of maternal depression, and you have 10 variables with 3 categories (30 categories), that means you are going to have too many of the cells in the table with very few or 0 cases, and this makes the estimates unstable.
Analyses will be intention to treat but the sections dealing with missing data, compliance-based analyses and other complex issues require more explanation. The multiple regression models appear to be using a form of forward stepwise regression without attention to causal relations of variables or questions of total causal effect versus effect decomposition into direct and indirect effects. There are planned subgroup analyses for number of CBT sessions, anti-depressant use, symptom clusters, severity of DASS21, additional health services, participant characteristics, mental health history and gestational age. These represent a mix of baseline and post-randomization variables. It is not clear how these results would be interpreted, which is important given that some of the variables appear to be affected by the intervention itself. As such, they appear more related to the structural equation modeling described in the Mechanisms section and it is not clear why they are included in this section at all.
Phase 2 methods are briefly described as semi-structured interviews, audio-taped and transcribed. Section 5.3 describes coding and analysis in 8 lines.

Feasibility
The applicants say 60 pregnant women per week are seen of which they expect to have ~10 eligible and 4 to participate. The study requires a research coordinator to be onsite. This means the research coordinator is in the physicians' office 3 ½-days per week to recruit 4 patients per week. The online questionnaires would facilitate data collection and analysis, but also includes sensitive information. The applicants say attrition rates for the online CBI program are half of group CBT.