Clinician self-efficacy in initiating discussions about gestational weight gain

Abstract

Objective

To compare prenatal care providers’ perceived self-efficacy in starting discussions about gestational weight gain with pregnant women under a variety of conditions of gradated difficulty, when weight gain has been in excess of current guidelines.

Design

A 42-item online questionnaire related to the known barriers to and facilitators of having discussions about gestational weight gain.

Setting

Canada.

Participants

Prenatal care providers were contacted through the Family Medicine Maternity Care list server of the College of Family Physicians of Canada.

Main outcome measures

The 42 items were clustered into categories representing patient factors, interpersonal factors, and system factors. Participants scored their self-efficacy on a scale from 0 (“cannot do at all”) to 5 (“moderately certain can do”) to 10 (“highly certain can do”). The significance level was set at α = .05.

Results

Overall, clinicians rated their self-efficacy to be high, ranging from a low mean (SD) score of 5.14 (3.24) if the clinic was running late, to a high mean score of 8.97 (1.34) if the clinician could externalize the reason for undertaking the discussion. There were significant differences in self-efficacy scores within categories depending on the degree of difficulty proposed by the items in those categories.

Conclusion

The results were inconsistent with previous studies that have demonstrated that prenatal care providers do not frequently raise the subject of excess gestational weight gain. On the one hand providers rate their self-efficacy in having these discussions to be high, but on the other hand they do not undertake the behaviour, at least according to their patients. Future research should explore this discrepancy with a view to informing interventions to help providers and patients in their efforts to address excess gestational weight gain, which is increasingly an important contributor to the obesity epidemic.

Excess weight gain in pregnancy is a risk factor for a number of adverse outcomes for mothers and their offspring, including downstream childhood obesity.^1–3 Guidelines exist pertaining to how much weekly and total weight should be gained based on women’s prepregnancy body mass index (BMI) category.^4,5 In Canada, more than 50% of women gain weight in excess of these guidelines.^6–8

A number of factors influence the guideline concordance of gestational weight gain, including advice from a prenatal care provider.^9–11 Although research shows that patients want their primary care clinicians to address gestational weight gain,¹² such discussions tend to occur infrequently,^11–18 especially as perceived by patients. Clinicians consider these conversations to be of a sensitive nature, and they fear offending, angering, or embarrassing their patients by raising the topic.^13,19–23 They have been shown to be more likely to provide advice about gestational weight gain when women are nulliparous, have a higher education, or have a higher socioeconomic status (SES).^9,11,24 They are also more likely to provide advice if they can externalize the discussion toward the health of the baby¹⁹ or toward weight-related comorbidities.¹¹ Prepregnancy BMI does not appear to influence the provision of advice, and there is controversy about the effect of maternal age.^9,11,24,25

Anecdotal evidence suggests that clinicians perceive few barriers to briefly addressing gestational weight gain when the gain has been congruent with the guidelines. When the weight gain has been incongruent with the guidelines, they perceive less discomfort in raising the issue in the context of inadequate compared with excess weight gain. As most women gain weight in excess of the guidelines, this is potentially a concern.

Previous studies have shown that clinician-perceived barriers to initiating discussions about gestational weight gain include a lack of time^15,26,27 and a lack of confidence.^28,29 From a theoretical perspective, confidence is related to perceived self-efficacy. This latter construct refers to “judgments of how well one can execute courses of action required to deal with prospective situations.”³⁰ According to Albert Bandura, a psychologist renowned for his contribution to this field, such judgment is mediated by 4 sources of information,³¹ perhaps the most important one being performance outcomes, or mastery experiences, which refers to the influence of positive or negative past experiences on one’s ability to perform a task. Bandura further states that there is no “all-purpose” measure of self-efficacy³²—a given individual might have high self-efficacy in some domains and low self-efficacy in others, and variations within a domain occur with diverse levels of difficulty within that domain.

The aim of this study was to compare clinicians’ perceived self-efficacy in initiating discussions about gestational weight gain with pregnant women under a variety of conditions and in the context of weight gain in excess of current guidelines.

METHODS

Approval for this study was obtained from the Nova Scotia Health Authority Research Ethics Board in Halifax, NS. Survey items that relate to the known barriers to and facilitators of having discussions about gestational weight gain were created consistent with Bandura’s approach. Before data collection, these items were proofread and examined for face validity by the author’s workplace colleagues who provide regular prenatal care. The final list of 42 items was entered into a Web-based survey system. Potential participants (N = 91) were reached via e-mail through contact information in the Family Medicine Maternity Care list server of the College of Family Physicians of Canada. Completion of the questionnaire constituted informed consent. Participants’ professions were purposefully not explored in order to reduce the chance of bias owing to the rumoured, albeit unsubstantiated, existence of “turf wars.”^33–35 Following a period of 8 weeks, the data were extracted from the survey software into SPSS, version 21.

The 42 items represented various conditions and were clustered into a number of categories. These categories included patient factors such as patient prepregnancy BMI, age, parity, education, SES, and presence of weight-related comorbidities; interpersonal factors such as clinician familiarity with the patient, understanding of the patient’s culture, perceived quality of the patient-clinician relationship, comparison of the patient’s prepregnancy BMI to the clinician’s own BMI, and whether the patient attended the prenatal visit alone or accompanied by someone; and clinic factors such as the degree to which the clinic was running on time and time allotted for the prenatal visit. For each item, participants were asked to rate their self-efficacy in starting a discussion about gestational weight gain with a patient who had gained more than the recommended amount on a scale from 0 (“cannot do at all”) to 5 (“moderately certain can do”) to 10 (“highly certain can do”). The significance level was set at α = .05. As the variables were not normally distributed (ie, they were skewed to the left) and were not independent, nonparametric tests (ie, Friedman tests and Wilcoxon signed rank tests) were used to compare the mean self-efficacy scores within the categories and between items.

RESULTS

Seventy-one clinicians completed the questionnaire, corresponding to a 78% response rate. Most respondents were female (94%); as a group, respondents had been providing prenatal care for a range of years, with approximately two-thirds of respondents reporting having provided prenatal care from 0 to 20 years (Figure 1).

Figure 1.

Length of time providing prenatal care

Self-efficacy was highest (mean [SD] 8.97 [1.34]) if the clinician could externalize the reason for undertaking the discussion, ie, if the patient had existing weight-related comorbidities (Table 1). Conversely, self-efficacy was lowest (mean [SD] 5.14 [3.24]) with system issues, ie, if the clinic was running 60 minutes late.

Table 1.

List of items in each category and mean self-efficacy scores: Self-efficacy was scored on a scale of 0 to 10, with higher scores indicating greater self-efficacy.

CATEGORY	ITEMS	MEAN (SD) SELF-EFFICACY SCORE
Patient’s prepregnancy BMI	Her prepregnancy BMI was in the underweight range	8.31 (2.12)
	Her prepregnancy BMI was in the normal range	8.53 (1.89)
	Her prepregnancy BMI was in the overweight range	8.35 (1.91)
	Her prepregnancy BMI was in the obese range	8.48 (1.94)
Patient’s age	She is a teenager	8.20 (2.14)
	She is a young adult	8.44 (2.02)
	She is approaching middle age	8.66 (1.60)
Patient’s parity	It is her first pregnancy	8.66 (1.87)
	She has had ≥ 1 previous pregnancies	8.28 (2.14)
Patient’s education	Her education ended in high school	8.42 (2.09)
	She obtained postsecondary education	8.63 (1.69)
	She has a postgraduate or professional degree	8.46 (1.82)
Patient’s SES	She is of low SES	8.17 (2.28)
	She is of middle SES	8.46 (1.98)
	She is of high SES	8.69 (1.82)
Comorbidities	She has 1 weight-related comorbidity	8.90 (1.35)
	She has > 1 weight-related comorbidity	8.97 (1.34)
Familiarity with the patient	I know her well	8.56 (1.87)
	I know her somewhat	8.18 (2.00)
	I do not know her at all	7.63 (2.56)
Understanding of patient’s culture	I understand her cultural background	8.82 (1.53)
	I do not understand her cultural background	6.89 (2.59)
Perceived quality of the clinician-patient relationship	I believe the quality of my relationship with her to be above average	8.92 (1.39)
	I believe the quality of my relationship with her to be average	8.33 (1.81)
	I believe the quality of my relationship with her to be below average	6.58 (2.57)
Attendance alone or accompanied	She attends the prenatal visit by herself	8.83 (1.46)
	She attends the prenatal visit accompanied by someone with an underweight BMI	7.66 (2.37)
	She attends the prenatal visit accompanied by someone with a normal BMI	7.87 (2.21)
	She attends the prenatal visit accompanied by someone with an overweight BMI	7.75 (2.29)
	She attends the prenatal visit accompanied by someone with an obese BMI	7.73 (2.34)
Comparison of patient’s prepregnancy BMI to clinician’s own BMI	Her prepregnancy BMI was a lot lower than mine	8.57 (2.10)
	Her prepregnancy BMI was lower than mine	8.53 (2.06)
	Her prepregnancy BMI was the same as mine	8.56 (2.04)
	Her prepregnancy BMI was higher than mine	8.32 (2.22)
	Her prepregnancy BMI was a lot higher than mine	8.24 (2.33)
Degree to which the clinic is running on time	My clinic is running on time	8.72 (1.75)
	My clinic is running 30 min late	6.21 (2.64)
	My clinic is running 60 min late	5.14 (3.24)
Time allotted for prenatal visit	I have 30 min for the visit	8.80 (1.50)
	I have 20 min for the visit	7.86 (2.10)
	I have 15 min for the visit	6.66 (2.59)
	I have 10 min for the visit	5.70 (3.04)

BMI—body mass index, SES—socioeconomic status.

There was no statistically significant difference in clinician self-efficacy based on the patient’s prepregnancy BMI or level of education. However, analyses for the other categories showed significant results. Friedman tests were used to compare items within categories containing more than 2 items (Table 2). Wilcoxon signed rank tests were used to compare pairs of items within the categories (Table 3).

Table 2.

Comparison of self-efficacy scores within categories using the Friedman test: Self-efficacy was scored on a scale of 0 to 10, with higher scores indicating greater self-efficacy.

CATEGORY	ITEMS	MEAN (SD) SELF-EFFICACY SCORE	FRIEDMAN TEST
Patient’s prepregnancy BMI	Underweight range	8.31 (2.12)	${\chi }_3^2=2.004$ P = .572
	Normal range	8.53 (1.89)
	Overweight range	8.35 (1.91)
	Obese range	8.48 (1.94)
Patient’s age	Teenager	8.20 (2.14)	${\chi }_2^2=13.698$ P = .001
	Young adult	8.44 (2.02)
	Approaching middle age	8.66 (1.60)
Patient’s education	High school	8.42 (2.09)	${\chi }_2^2=5.393$ P = .067
	Postsecondary	8.63 (1.69)
	Postgraduate or professional degree	8.46 (1.82)
Patient’s SES	Low SES	8.17 (2.28)	${\chi }_2^2=21.219$ P < .001
	Middle SES	8.46 (1.98)
	High SES	8.69 (1.82)
Familiarity with the patient	Familiar	8.56 (1.87)	${\chi }_2^2=16.958$ P < .001
	Somewhat familiar	8.18 (2.00)
	Not at all familiar	7.63 (2.56)
Perceived quality of the clinician-patient relationship	Above average	8.92 (1.39)	${\chi }_2^2=89.127$ P < .001
	Average	8.33 (1.81)
	Below average	6.58 (2.57)
Attendance alone or accompanied	Alone	8.83 (1.46)	${\chi }_4^2=56.677$ P < .001
	Companion with underweight BMI	7.66 (2.37)
	Companion with normal BMI	7.87 (2.21)
	Companion with overweight BMI	7.75 (2.29)
	Companion with obese BMI	7.73 (2.34)
Comparison of patient’s prepregnancy BMI to clinician’s own BMI	A lot lower	8.57 (2.10)	${\chi }_4^2=10.397$ P = .034
	Lower	8.53 (2.06)
	The same	8.56 (2.04)
	Higher	8.32 (2.22)
	A lot higher	8.24 (2.33)
Degree to which the clinic is running on time	On time	8.72 (1.75)	${\chi }_2^2=93.391$ P < .001
	30 min late	6.21 (2.64)
	60 min late	5.14 (3.24)
Time allotted for prenatal visit	30 min	8.80 (1.50)	${\chi }_3^2=127.275$ P < .001
	20 min	7.86 (2.10)
	15 min	6.66 (2.59)
	10 min	5.70 (3.04)

BMI—body mass index, SES—socioeconomic status.

Table 3.

Comparison of pairs of items within categories using the Wilcoxon signed rank test

CATEGORY	ITEMS COMPARED	Z STATISTIC	P VALUE
Patient’s age	Teenager vs young adult	−3.305	.001
	Teenager vs approaching middle age	−2.840	.005
	Young adult vs approaching middle age	−1.378	.168
Patient’s parity	Nulliparous vs primiparous or multiparous	−2.741	.006
Patient’s SES	Low SES vs middle SES	−3.086	.002
	Low SES vs high SES	−3.472	.001
	Middle SES vs high SES	−2.801	.005
Comorbidities	1 weight-related comorbidity vs > 1 weight-related comorbidity	−0.462	.644
Familiarity with the patient	Familiar vs somewhat familiar	−1.714	.087
	Familiar vs not at all familiar	−2.554	.011
	Somewhat familiar vs not at all familiar	−3.082	.002
Understanding of patient’s culture	Yes vs no	−5.618	< .001
Perceived quality of the clinician-patient relationship	Above average vs average	−4.030	< .001
	Above average vs below average	−6.180	< .001
	Average vs below average	−6.017	< .001
Attendance alone or accompanied	Alone vs companion with underweight BMI	−4.654	< .001
	Alone vs companion with normal BMI	−4.075	< .001
	Alone vs companion with overweight BMI	−4.467	< .001
	Alone vs companion with obese BMI	−4.467	< .001
	Companion with underweight BMI vs with normal BMI	−2.724	.006
	Companion with underweight BMI vs with overweight BMI	−0.535	.592
	Companion with underweight BMI vs with obese BMI	−0.284	.776
	Companion with normal BMI vs with overweight BMI	−1.174	.240
	Companion with normal BMI vs with obese BMI	−1.186	.235
	Companion with overweight BMI vs with obese BMI	−0.333	.739
Comparison of patient’s prepregnancy BMI to clinician’s own BMI	A lot lower vs lower	0.000	> .99
	A lot lower vs the same	0.000	> .99
	A lot lower vs higher	−1.834	.067
	A lot lower vs a lot higher	−2.107	.035
	Lower vs the same	−0.447	.655
	Lower vs higher	−1.563	.118
	Lower vs a lot higher	−1.903	.057
	The same vs higher	−2.025	.043
	The same vs a lot higher	−2.358	.018
	Higher vs a lot higher	−2.310	.021
Degree to which the clinic is running on time	On time vs 30 min late	−6.422	< .001
	On time vs 60 min late	−6.276	< .001
	30 min late vs 60 min late	−4.111	< .001
Time allotted to prenatal visit	30 min vs 20 min	−4.667	< .001
	30 min vs 15 min	−5.872	< .001
	30 min vs 10 min	−6.296	< .001
	20 min vs 15 min	−5.300	< .001
	20 min vs 10 min	−5.333	< .001
	15 min vs 10 min	−4.507	< .001

BMI—body mass index, SES—socioeconomic status.

Pertaining to patient factors, clinician self-efficacy scores were significantly higher for older patients compared with younger patients (P = .001), when patients were nulliparous compared with primiparous or multiparous (P = .006), and when patients had a higher compared with a lower SES (P < .001). There was no significant difference between the presence of 1 or several weight-related comorbidities.

When interpersonal factors were considered, self-efficacy scores were significantly higher with greater clinician knowledge of the patient (P < .001), when the clinician understood the patient’s cultural background compared with when there was no such understanding (P < .001), and with higher clinician rating of the quality of the clinician-patient relationship (P < .001). Clinicians rated their self-efficacy to be highest when a patient attended the prenatal visit by herself compared with when she was accompanied by anyone else. Self-efficacy scores tended to be lower when clinicians perceived the patient’s prepregnancy BMI to be higher than their own BMIs.

With regard to clinic factors, self-efficacy scores were significantly higher when the clinic was running on time versus running late, and there were significant differences in self-efficacy scores depending on the time available for the prenatal appointment (P < .001 for both), with more time being associated with higher scores. Notably, even the comparison of a 15-minute versus a 10-minute appointment was associated with significantly higher self-efficacy scores (P < .001).

DISCUSSION

While a number of factors have been identified as increasing or decreasing the likelihood that clinicians would raise the issue of excess gestational weight gain with their patients, to the author’s knowledge, this is the first theory-driven study to explore clinicians’ perspectives of how various scenarios affect their perceived self-efficacy in raising the subject.

Outside of prenatal care, there have been a number of interventions to address some of the barriers identified by clinicians wishing to have weight-related discussions with their patients, including training in behaviour change counseling^36,37 and the use of point-of-care tools.³⁸ Such interventions have shown some promise in terms of patient health behaviour and health outcomes,^39,40 and in particular, have demonstrated a 2-fold increase in clinician initiation of weight-related discussions with their patients.⁴¹

With the recent increased attention on gestational weight gain and its short- and long-term health implications for both mothers and their children, it is prudent to translate this momentum to the prenatal care period, a period that has been dubbed a “teachable moment.”⁴² However, it is not sufficient to focus solely on behaviour change initiatives that target patients—efforts should also be directed toward prenatal care providers. Previous studies have shown an inconsistency between patients’ and clinicians’ reports of whether discussions about gestational weight gain actually take place.^13,15 Perhaps this is partly owing to the manner in which these discussions are initiated and conducted,⁴³ and interventions as described above could address this problem. The Canadian Obesity Network recently launched a tool for prenatal care providers, The “5 As of Healthy Pregnancy Weight Gain,”⁴⁴ the development of which was informed by the patient-centred clinical method⁴⁵ and behaviour change theory. At the core of this tool is the philosophy that advice provided to patients should be given in the context of having first understood the whole person (ie, the patient’s context) so that common ground can be found. Led by behaviour change specialists, current work is under way in a number of regions throughout Canada to develop standardized training modules relevant to this tool and to evaluate the effect of training clinicians in the use of this tool. This study adds to the literature and will inform this endeavour by providing the clinicians’ perspectives on where their challenges lie pertaining to having discussions about gestational weight gain with their patients.

Limitations

Although clinician self-efficacy in discussing gestational weight gain with their patients was clearly vulnerable to system issues and to the patient-clinician relationship, it was nonetheless fairly high overall. This is somewhat surprising considering previous work demonstrating that prenatal care providers do not frequently raise the subject with their patients and that they perceive such discussions to be difficult owing to the generally sensitive nature of weight-related discussions. This reflects a potential weakness of the study—perhaps the participants were not truly representative of prenatal care providers in general, as they had self-selected to be part of the College of Family Physicians of Canada’s maternity list server and were therefore likely to be providers of low-risk prenatal care with an interest in prenatal care. Assuming this to be true, the current study would reflect the best-case scenario in terms of self-efficacy scores, but the relative self-efficacy scores for items within categories would likely persist.

Conclusion

Understanding the perspectives of prenatal care providers is an important step for clinicians, researchers, and decision makers interested in developing interventions that are relevant to these clinicians and address their concerns. Such undertakings could likely benefit not only the clinicians, but ultimately the women and families for whom they provide clinical care.

EDITOR’S KEY POINTS

• This is the first theory-driven study to explore clinicians’ perspectives of how a variety of scenarios affect their perceived self-efficacy in raising the subject of excess gestational weight gain.

• Self-efficacy was highest (mean [SD] 8.97 [1.34]) if the clinician could externalize the reason for undertaking the discussion, ie, if the patient had existing weight-related comorbidities. Conversely, self-efficacy was lowest (mean [SD] 5.14 [3.24]) with system issues, ie, if the clinic was running 60 minutes late.

• Although clinician self-efficacy in discussing gestational weight gain with their patients was clearly vulnerable to system issues and to the patient-clinician relationship, it was nonetheless fairly high overall. This is somewhat surprising considering previous work demonstrating that prenatal care providers do not frequently raise the subject with their patients and that they perceive such discussions to be difficult owing to the generally sensitive nature of weight-related discussions.

POINTS DE REPÈRE DU RÉDACTEUR

• Il s’agit de la première étude axée sur la théorie visant à explorer le point de vue des cliniciens sur la façon dont divers scénarios influent sur leur perception de leur efficacité personnelle à soulever le sujet d’un gain pondéral gestationnel trop élevé.

• La perception de sa propre efficacité était la plus élevée (moyenne [ET] de 8,97 [1,34]) si le clinicien pouvait exprimer la raison d’amorcer la discussion, par exemple, si la patiente avait déjà des comorbidités liées au poids. À l’inverse, la perception de sa propre efficacité était la plus basse (moyenne [ET] de 5,14 [3,24]) en présence de problèmes systémiques, par exemple, si la clinique accusait 60 minutes de retard dans les rendez-vous.

• Même si l’efficacité perçue par le clinicien luimême à discuter du gain pondéral gestationnel avec leurs patientes était clairement sensible aux problèmes du système et à la relation patientemédecin, elle était néanmoins plutôt élevée dans l’ensemble. Cette constatation est quelque peu surprenante compte tenu de travaux antérieurs faisant valoir que les professionnels qui offrent des soins prénatals soulèvent rarement ce sujet avec leurs patientes et ont l’impression que ces discussions sont difficiles en raison de la nature généralement délicate des conversations relatives au poids.