Problems Encountered in Single-center Prospective Studies in Pregnancy
Prospective research involving pregnant women is challenging for both the pregnant woman and the researcher (Box 1). In trials enrolling pregnant women, there are two patients involved: the mother and the fetus. In addition, the ethical considerations, informed consent requirements and risk–benefit issues are often more complex than in nonpregnant populations. In therapeutic prospective studies, these issues become much more complex, as the risks and benefits of proposed interventions are often more complex and may result in opposite outcome for the mother and the fetus. Difficult enrollment and high drop-out rates are inherent in pregnancy studies (Box 2). A few prospective randomized studies in pregnancy are ongoing and have published study designs, and their final results are awaited. The investigators of a prospective study that evaluates the role of exercise programs in pregnant women have observed adherence of women to exercise in the exercise group, which is challenging given the patients' sedentary habitus and is compounded by obesity and pregnancy. The publication of this study's results is expected in 2016 [MOHOLDT T, PERS. COMM.].
Research Mandates, Institutional Factors & Informed Consent
Although pregnant women have the capacity to make autonomous decisions, they are considered to be vulnerable due to potential harm to the fetus. In the USA, the NIH mandates for the inclusion of racial/ethnic groups in order to encourage accountability in research design and federal funding. The Department of Health and Human Services (DHHS) requires that investigators must balance their interest in gathering data and answering research questions with society's mandate to protect the rights of, and safeguard the welfare of, research participants. The Institutional Review Board requires that the risk is no greater than minimal and that the purpose of the research is for the development of important biomedical knowledge that cannot be obtained by other means. The investigator has to keep this in perspective and adhere to these mandates when working with obstetric participants, while simultaneously developing successful strategies for the recruitment and retention of research participants (Box 3). Many barriers in academic institutions, including the research time that is available to faculty and the lack of interdepartmental collaboration, as highlighted in Box 4, are barriers that investigators face when conducting research into pregnancy.
In order to balance federal mandates against the ability of the patient to easily comprehend consent forms, these consent forms should be clearly written (limiting the amount of text while meeting the federal and privacy regulations) and have the potential to increase the participants' understanding of their privacy rights, without sacrificing the comprehension of the research study. The researchers are responsible for explaining why the study is being conducted, what it involves and the harms and benefits that are associated with the research. A potential participant has the right to clear and full explanations regarding the research and answers to any and all questions. The purposes, risks and benefits of the study should be explained without overwhelming the potential participant. Tailoring the delivery of information to each specific client is also helpful.
Study Participation & Retention
Since many physiologic and anatomic changes occur at as early as at 6–7 weeks of gestation, obtaining baseline data once pregnancy has occurred is less than ideal. Thus, in order to obtain control values, postpregnancy data are collected. After pregnancy, a number of these women may experience life events prohibiting further participation (e.g., multiple gestation, an unexpected move, lack of childcare for their newborn, post-partum depression, post-partum physical pain or another pregnancy, among others). Most participants may be reluctant to commit to studies or may enroll underestimating the commitment involved in participating. Data collection prior to pregnancy offers a more complete picture of changes associated with the entire gestational period and beyond. While optimal from a research perspective, this strategy adds further challenges to recruitment and retention. Longitudinal studies involving prepregnancy data collection have the added challenge of the unpredictability of conception and finite gestation, limiting recruiting flexibility and scheduling. Prior studies show that if 100 women actively trying to conceive are recruited as study participants, statistically, only 80% will become pregnant over a 12-month period, and 20–30% of these women will spontaneously miscarry, reducing the number of research participants for follow-up to as low as 64. More recently, however, the feasibility of conducting such studies has been demonstrated. Of 143 women, 101 women conceived within 18 months and 70 had live births' three women developed pre-eclampsia and two had intrauterine growth restriction. Two were lost to follow-up. Enrolling women in the first trimester of pregnancy is difficult. First-trimester women may not know they are pregnant, may not receive any prenatal care for the first few months and often present to the clinic after 14 weeks of gestation. In order to obtain baseline data, the post-partum state is evaluated. However, when using post-partum as the baseline period, it is important to remember that several changes take time to return to their true baseline values, and some may not return to normal at all. Studies vary in the timing of echocardiograms, from as early as 2–4 weeks post-partum to 6 months post-partum. True preconception baseline values are the optimal, but in the absence of this, the 6-month time-point appears to be the best surrogate, although while the hemodynamic changes may have returned to preconception values by 6–12 weeks, structural changes can take months to return to normal. Differences in the timing of the post-partum baseline values probably account for some of the discrepancies in data between pregnancy studies.
Reaching out to the study participants is another challenge. An element of coercion is considered inherent if an investigator approaches his/her own clinical patients for enrollment; in the research study, the investigator is both a clinician as well as a researcher. The study subjects should preferably be approached by noninvestigators in a neutral manner and should be given ample time to consider research participation with the reassurance of no change in clinical care should she decide not to participate in the research study.
Given restrictions on approaching patients (especially one's own patients), having the patients approach the investigator regarding the study in response to advertisements (via newspapers, radio, email or posts in patient waiting areas) regarding the research study minimizes the element of coercion in research study participation. However, such strategies are expensive, require dedicated budgets and often may need to be repeated in order to encourage participation. The most common methods of advertisement are newspaper and radio adverts. For smaller-scale adverts, emails, study flyers, grand rounds and lectures, among others, are used in order to inform other physicians regarding research studies. Among targeted emails, flyers and newspaper adverts, the most successful advertising strategy for both recruitment and retention is local newspaper adverts. The CDC suggests a 70% retention rate for confidence in validity.
Budget & Funding
Research designs should aim to include procedures and interventions that clinicians can deliver, given the typical demands of practice in community settings, as well as in academic medical centers. This will reduce clinical physician and staff resistance to conducting research studies. It is essential to create an ample budget that allows for study advertisement, follow-up adverts and the salaries of the research staff needed in order to ensure adequate participant recruitment and retention. Supplemental funds, such as hospital intramural grants and external university summer grants, should be considered. Investigators should consider paying medical students to assist research staff in the recruitment of participants and employing additional clerical staff in order to enhance retention capabilities throughout the study. Sharing of the equipment needed for research studies, in addition to sharing of the research space with the clinical space, will help to reduce the study costs (Box 5). Administrative and institutional commitment and support for pregnancy research is necessary in order to have shared resources (Box 4). Box 6 lists items to consider when incorporating imaging studies, such as ultrasound, during pregnancy. Table 1 summarizes the problems that are encountered in pregnancy research and their potential solutions.