Antenatal physical activity counseling among healthcare providers ()
1. INTRODUCTION
Despite the fact that the American Congress of Obstetricians and Gynecologists (ACOG) recommends pregnant women who are free of obstetrical complications engage in leisure-time physical activity (LTPA), women who are pregnant engage in less LTPA than their nonpregnant counterparts [1,2]. Specifically, Evenson et al. 2004 found that only 15.1% of pregnant women were engaging in the recommended amounts of LTPA versus 26.1% of nonpregnant women. Moreover, LTPA levels often decrease throughout gestation [3]. Given the lower rates of LTPA among pregnant women, coupled with the many maternal benefits associated with antenatal LTPA such as increased fitness [4], prevention of excess weight gain [5], low back pain [6], and lower anxiety and depressive symptomatology [7], there is a definite need to promote engagement in antenatal LTPA [3,8].
Pregnancy is an opportune time for women to modify their health behaviors. This time can be thought of as an “external motivator” to elicit behavior change in order to protect the health of the fetus as well as the mother’s own health. Providing pregnant women with the necessary knowledge to promote engagement in physical activity can not only positively impact antenatal health outcomes (e.g. excessive antenatal weight gain, antenatal mood disorders, among others) but also those that occur in the postpartum period and beyond [9].
Engaging in regular physical activity decreases an individual’s risk of overweight and associated chronic diseases such as, cardiovascular disease, type 2 diabetes mellitus, osteoporosis, obesity, and certain types of cancer (see review [10]). In particular, participation in walking (3 times a week for 45 min or 5 times a week for 30 min) is linked to reduced risk of coronary events in women [11]. Thus, promoting physical activity during pregnancy may not only reduce the risk of excessive antenatal weight gain, but also reduced risk of future overweight and chronic disease.
Given the demonstrated benefits of antenatal physical activity, why do many women fail to engage in LTPA? Pregnant women often report a lack of knowledge concerning the safety of exercising during pregnancy and believe that if they received information related to how to safely and effectively exercise during pregnancy it would facilitate their engagement in physical activity [12-14]. Incorrect information or complete absence of counseling from healthcare providers (HCPs) may compound this problem. This is unfortunate, given that brief physical activity counseling by HCPs increases the likelihood that patients will engage in physical activity [15,16]. Moreover, recent research suggests that the majority of women indicate that their HCPs have the most influence on their beliefs regarding physical activity [16]. Even with clear evidence delineating the benefits of antenatal physical activity counseling by HCPs on numerous maternal and child health outcomes, much of the healthcare system has not promoted effectively the concept that LTPA can be used to prevent and treat disease.
To date, there is a dearth of literature examining the extent to which HCPs provide brief counseling on antenatal physical activity to their pregnant patients. Much of the extant literature has been narrow in scope and limited to small, convenience samples [17-19] or examines primary care systems outside of the US, such as Australia [20]. For example, Entin and Munhall administered an 18-item survey to 83 obstetricians in private or small group practices in the US. Approximately half of the obstetricians surveyed reported that they do not routinely discuss exercise with their patients, and that most are hesitant to advise sedentary women to start exercise [17]. Bauer et al. (2004) surveyed 60 practicing physicians in Michigan and found that the majority believe exercise is beneficial and that they recommend exercise to their patients; however, it appears that not all are aware, or follow, current ACOG recommendations [18]. Recently Hughes et al. 2011 surveyed HCPs in general practice, community nursing and pharmacy to determine continuing education needs related to perinatal physical activity counseling. Results suggested that the majority of providers agree that physical activity guidance is important and recommend that future continuing education efforts be tailored according to type of primary care training [20]. Moreover, a large surveillance study which focused primarily on obesity prevention and weight gain counseling among primary care HCPs in the US provides some insight as to prevalence rates of antenatal physical activity counseling but fails to provide a comprehensive assessment of the potential predictors of counseling such as HCPs’ attitudes, beliefs and perceived barriers [21]. A more comprehensive assessment of change agents (e.g., attitudes, beliefs, perceived barriers) would provide the critical information needed to inform future intervention efforts designed to promote antenatal physical activity counseling among HCPs. The present study addresses this gap in the literature by providing a more comprehensive assessment of potential factors (i.e. beliefs, attitudes, knowledge, self-efficacy, barriers) related to antenatal physical activity counseling among a diverse sample of HCPs who provide obstetric care. Data obtained from this study will help inform future interventions aimed at enhancing antenatal physical activity counseling in primary care settings.
2. MATERIALS AND METHODS
Prior to conducting this investigation, the study protocol was approved by the Colorado Multiple Institutional Review Board. A HCP was eligible to participate in the study if he/she was a physician (i.e., practicing in obstetrics or family medicine) or a certified nurse midwife (CNM) who was currently practicing in the DenverAurora Metropolitan Statistical Area (DAMSA). Eligible HCPs who met the study inclusion criteria were identified through local hospital and clinical directories and local chapters of the American Academy of Family Medicine, Certified Nurse Midwives, and the American College of Obstetricians and Gynecologists. Our initial search yielded 1210 names and included all MDs practicing obstetrics, all family doctors, and CNMs in the DAMSA. From this list, 59 obstetricians (OBs) and 7 CNMs were removed from the sample due to incorrect contact information. Additionally, 569 family medicine physicians (FMP) were eliminated due to either not providing obstetric care for their patients (54%) or incorrect current contact information (19%).
A total of 575 HCPs (i.e. 259 OBs; 104 CNMs, and 212 FMPs) were invited to participate in the study. A pre-notification of the study was sent via email or facsimile to all eligible HCPs to increase awareness of the upcoming invitation to participate in the survey. Within one week of the pre-notification, each eligible HCP was sent an email and/or facsimile that briefly described the survey and that contained a web link to the site where the survey would be hosted. In cases in which a HCP did not have access to email or fax, a paper invitation (containing the web address to the survey) was mailed to the individual via regular postal service. Additional followup attempts were made by email and postal mail notifications by the research team as well as by the HCPs respective organizations including the Colorado section of ACOG, Colorado section of the American College of Nurse Midwives, and the Colorado Academy of Family Physicians. Participants were given the opportunity to complete the survey by mail as well as by web. The majority of respondents (i.e. 55%) responded by paper survey.
The final sample comprised 188 HCPs currently practicing in one of three specialties (i.e. family medicine, obstetrics, and midwifery) in the DAMSA. The sample included 91 obstetricians, 40 midwives, and 57 family medicine physicians, resulting in an overall response rate of 32.7%. There were no significant differences between responders and nonresponders on age and county in which the HCP worked. However, there was a significant difference in response rate by gender among obstetricians only , with female obstetricians more likely than male obstetricians to respond (p = .014). Demographics for the full sample and by specialty are shown in Table 1.
Table 1. Sample description by specialty.
Steps were taken to ensure confidentiality of respondents who completed the survey on the internet. Study data were collected and managed using the Research Electronic Data Capture System (REDCap). REDCap is a secure, web-based application designed to support data capture for research studies [22]. Prior to completing the web-based survey, participants were presented with an information page which provided the informed consent form describing the purpose of the study, criteria for participation, confidentiality measures, incentive details, and contact information for the principal investigator and study coordinator. Participants were also reminded that their responses pertained to only pregnant women who were free of obstetric complications. Agreement to participate was confirmed by clicking on a “Continue” button which directed users to the survey. After completing the survey, participants were redirected to a separate, secure web page where they entered their name and e-mail address; this information was needed in order to receive the incentive (a $15 gift card). Contact information was stored in a separate database such that it could not be linked to survey responses. Individuals who elected to complete the paper survey, provided a written informed consent and data were then entered using double data entry into the REDCap database.
2.1. Data Management and Analyses
Given the proposed comprehensive assessment, a survey was developed to assess HCPs’ beliefs, attitudes, knowledge, and self-efficacy related to the counseling of antenatal physical activity. Attitudes toward antenatal physical activity focused on the physician’s perceived level of responsibility toward providing counseling on antenatal physical activity (e.g. “It is an important responsibility of mine to discuss physical activity with my pregnant patients”). Physician beliefs related to benefits associated with antenatal physical activity (e.g. “Exercise during pregnancy improves the health for the mother”), their beliefs related to the safety of exercise during pregnancy (e.g. “Pregnant women should decrease their exercise as their pregnancy progresses”), and their beliefs related to who should exercise during pregnancy (e.g. “A sedentary woman, with an uncomplicated pregnancy, should not begin an exercise program during pregnancy”). The construct assessing the HCP’s level of efficacy toward providing counseling on antenatal physical activity included items such as “I am confident in my knowledge regarding the benefits and risks of physical activity during pregnancy” and “I am confident in my ability to effectively talk with pregnant patients about antenatal physical activity”. HCP knowledge of antenatal physical activity pertained to their knowledge of the 2002 American Congress of Obstetricians and Gynecologists and the 2008 Department of Health and Human Services guidelines.
Perceived barriers toward providing antenatal physical activity counseling were also assessed (e.g. “To what extent does: 1) insufficient time; 2) inadequate knowledge; and 3) skills/uncertainty about available resources prevent you from administering physical activity counseling to your pregnant patients”). Current practices were captured by determining the percentage of patients who HCPs currently provide antenatal physical activity counseling to and in what form this is delivered: information (e.g. brochures, pamphlets, handouts, etc.), referrals, or in-office counseling by HCP or office staff. In addition, demographic and practice site information were collected such as age, gender, race, type of degree, type of healthcare provider, specialty area, years of practicing, and practice location (urban, suburban, rural). Last, we also assessed if and how healthcare providers would like future training related to antenatal physical activity counseling.
Recommended guidelines for developing survey questionnaires were utilized [23]. First, in order to create the survey an initial pool of approximately 75 items assessing the aforementioned constructs were developed. Some of the proposed items also were extracted from a preexisting survey “Exercise and Pregnancy Survey for Healthcare Providers” [18]. Constructs were conceptualized into two general classes which were treated differently throughout the survey development process. SingleItem traits were constructs that were presumed to be measured without error and which are only represented by a single item on the survey. Examples of single-item traits included sex, race, practice type, and other similar variables. Multi-item latent traits were measured with multiple survey items that were all presumed to measure an underlying trait with some measurement error. Examples of multi-item latent constructs include attitudes, beliefs, perceptions, knowledge, and general classes of behavior.
An expert panel of 11 HCPs (i.e. 3 OBs, 5 CNMs, 3 FMPs) were selected to perform the initial evaluation of the item pool. Panelists were asked to complete the online survey of 75 items and then asked to review, rate, and comment on all items in the initial pool. Participants were reminded before the onset of the survey that their responses pertained to only pregnant women who were free of obstetric complications. Panelists also provided qualitative feedback (e.g. wording changes, impressions, etc.) on individual items, survey sections, and the survey overall. Upon completion of this process, comments and feedback were transcribed and summarized for common themes. Single-item traits were revised based on commonalities across input from panelists. Any item that received a below-median average rating from panelists was rewritten or dropped from the item pool.
Multi-item traits were examined through standard item-scale diagnostics that included item distributional qualities, scale reliability, and item contributions to the scale (e.g., item-total correlations, reliability increases from each item). Items that were invariant, strongly skewed, showed prominent ceiling or floor effects, or which made negligible or negative contributions to scale reliability were dropped or rewritten. Panelist ratings of items were also considered in deciding item retention and drop decisions.
The final 39-item close-ended survey was then placed on a secured web-site designed for purposes of the proposed project. Consistent with current recommendations, the online survey was tested for user-friendliness [24]. To do so, a group of healthcare providers and researchers were asked to complete the web-based survey. The structure of the survey and whether it flows smoothly and in a logical manner, as well as survey completion time was assessed. During this pilot phase, written and verbal feedback was obtained and appropriate modifications were made resulting in a survey instrument that was self-administered in less than 10 minutes.
2.2. Analyses
Descriptive statistics for demographics and items for each construct were computed. One-way ANOVAs and chi-square tests of association were used, as appropriate, to explore potential differences on current healthcare provider characteristics (e.g. attitudes, beliefs, knowledge, current practices), and perceived barriers across the two specialties and healthcare provider types. All analyses were conducted using SAS Version 9.2.
3. RESULTS
Table 1 shows the descriptive characteristics of the study sample. More participants were trained as MDs specializing in obstetrics (48%) while family medicine physicians and certified nurse midwives comprised smaller proportions of the sample, 30% and 21%, respectively. The majority of the respondents were female (71%), white (92%), non-Hispanic (97%), and practicing in an urban setting (55%). Most participants had been providing healthcare for at least 10 years (78%) and the majority (64%) had been practicing in the DAMSA area for 5 years or more.