US Pediatricians’ Attitudes, Beliefs, and Perceived Injunctive Norms About Spanking

Abstract

Objective:

To assess United States pediatricians’ attitudes, beliefs, and perceived professional injunctive norms regarding spanking.

Method:

A self-administered questionnaire was mailed to a nationwide random sample of 1500 pediatricians in the US, drawn from a database maintained by IMS Health. Four survey mailings were conducted and cash incentives of up to $20 were provided. The response rate was 53% (N = 787).

Results:

Most respondents were members of the American Academy of Pediatrics (85%), had been practicing physicians for 15 years or more (66%), and were white (69%) and female (59%). All US regions were represented. About 3-quarters of pediatricians did not support the use of spanking, and most perceived that their colleagues did not support its use either. Pediatricians who were male, black, and/or sometimes spanked as children had more positive attitudes toward spanking and expected more positive outcomes from spanking than their counterparts. Nearly 80% of pediatricians never or seldom expected positive outcomes from spanking, and a majority (64%) expected negative outcomes some of the time.

Conclusion:

The majority of pediatricians in the US do not support the use of spanking with children and are aware of the empirical evidence linking spanking with increased risk of poor health outcomes for children. Pediatricians are a key, trusted professional source in advising parents about child discipline. These findings suggest that most pediatricians will discourage the use of spanking with children, which over time could reduce its use and associated harms in the population. This is of clinical relevance because, despite strong and consistent evidence of the harms that spanking raises for children, spanking is still broadly accepted and practiced in the US

Through empirical evidence amassed over the past several decades, a consensus has gradually been emerging about the harmfulness of hitting children for disciplinary purposes, including increased risks for child physical abuse, aggressive behavior, and negative parent-child relationships,¹ as well as mental and physical health problems into adulthood.^2–4 This practice goes by many names, most commonly spanking or corporal punishment (CP), which is defined as: “…the use of physical force with the intention of causing a child to experience pain but not injury for the purposes of correction or control of the child’s behavior (p.3).”⁵ Despite the emergent evidence of harm to children, the use of CP is still widely accepted and practiced in the United States, with 65% of women and 76% of men believing that CP is sometimes necessary and 80% using CP on their children by kindergarten.^6,7 The high prevalence of this practice in the US combined with its evidence of harm to children makes the widespread use of CP an important public health issue. Hence, efforts are underway to understand how best to change these norms and practices in the US as well as around the world.^8–11

The Theory of Planned Behavior¹² suggests that both expected outcomes and perceived norms of a behavior are often strongly linked with attitudes toward the behavior. Accordingly, parents’ attitudes toward using CP are strongly associated with their expected outcomes of CP use as well as their perceived norms of family, friends, and key professionals regarding CP use.¹³ In particular, when parents have been asked to identify the professional they are most likely to seek and heed advice from regarding parenting and child discipline, pediatricians, and medical professionals in general, have emerged among the most credible sources.^14,15

Few studies have assessed pediatricians’ attitudes regarding CP use. Data gathered 25 years ago showed that 59% of pediatricians, and 79% of family physicians in Ohio, approved of the use of CP despite the evidence of harm at the time.¹⁶ A few years later, in 1997 to 1998, the American Academy of Pediatrics (AAP) found through their periodic survey that a majority of pediatricians (53%) were mostly opposed to the use of CP by parents but believed it could sometimes be effective.¹⁷ This latter study is the only one we are aware of that aimed to obtain a representative sample of US pediatricians. More recent studies have assessed these attitudes among wider groups of medical professionals within highly targeted samples. One such study conducted at hospitals in Wisconsin and Missouri found that about 45% of medical staff agreed that spanking is harmful to children, yet about 87% agreed that spanking could lead to injury; however, only 4% of this sample was physicians.¹⁸ Nearly all of these physicians had witnessed parents hitting their child in the medical center at some point in the prior year; of those who had witnessed this behavior in the prior 3 months, only 20% never attempted to intervene.¹⁹ Those that did intervene usually spoke directly with the parent, were comfortable doing so, and felt that their intervention was effective. Another recent study of professionals’ attitudes toward CP targeted the membership of the “American Professional Society on the Abuse of Children”; physicians made up 21% of this highly specialized sample. A very high percentage of these professionals were opposed to CP (82% thought it was a bad disciplinary technique; 74% thought it was harmful to children), and attitudes toward CP did not differ by profession.²⁰ However, it is likely that professionals who do not specialize in child abuse have lower rates of opposition to CP than those found in this study.

To our knowledge, no other survey aside from the AAP’s 1997–98 study has assessed a representative sample of U.S. pediatricians or physicians for their views on this topic in the past 2 decades. Furthermore, no study has examined those same key variables that can shape attitudes about CP amongst parents—expected outcomes of CP use and perceived norms of professionals—amongst physicians. The purpose of this study is to fill these gaps and to assess US pediatricians’: (1) personal attitudes toward CP use, (2) perceptions of normative attitudes toward CP use amongst their colleagues, (3) beliefs about expected outcomes of CP use, and (4) demographic variations across these constructs.

METHODS

Study Procedure

The sampling frame for this study was drawn in February 2016 from a national database of US physicians who indicated “pediatrics” as their primary specialty (N = 33,476). The study population was comprised of 1500 randomly selected records purchased from SK&A (now IQVIA). The University of Wisconsin Survey Center obtained the sample and conducted survey mailings and data collection, using the main techniques suggested for maximizing survey response rates among health care professionals including mailed (versus online) surveys, monetary incentives, and multiple follow-ups.²¹ The survey design included 4 mailings: (1) a full mailing, which included a cover letter, a questionnaire, a cash incentive, and a business reply envelope (N = 1500); (2) a postcard reminder sent one week later (N = 1500); and (3) 2 additional full mailings 1 month (N = 947) and 7 weeks (N = 757) later to participants who had not yet responded. Items were mailed between March 25 and May 13, 2016. Final data collection ended on June 17, 2016. The following instructions were included in the full mailings:

I am writing to ask for your help with our survey of pediatricians. The aim of this study is to learn about your opinions, training needs and experiences, and professional practices regarding advising parents about child discipline and related parenting issues. We hope these findings will help inform efforts to improve child health outcomes.

Participants received anywhere from $5 to $20 in cash incentives across 2 mailings. An incentive treatment experiment was embedded into this study in order to assess the impact of variations in incentive administration, the results of which have been reported elsewhere.²² The survey was completed by 787 respondents for a final response rate of 53%. Our response rate is on par with results from a meta-analysis of health care professional surveys, which found an average response rate of 50%, with higher rates for mailed surveys with strong follow-ups and incentives.²¹ All responses remained anonymous. No personally identifying information was collected. The study was approved by the Institutional Review Board of the Tulane University Human Research Protection Program.

Measures

Attitudes toward corporal punishment

We used a shortened 4-item version of the attitudes toward spanking (ATS) questionnaire given the desired brevity of the survey instrument.²³ Items were selected based on recommendations from Dr. Holden (ATS author) as well as the high reliability scores obtained with this shortened scale in prior studies: α = 0.79¹³ and α = 0.81.²⁰ Participants were given the following instructions: “The following statements refer to your opinions about spanking children. For the purposes of this survey, spanking refers to hitting a child with a hand or an object with the intention of causing pain, but not injury, for the purpose of correction or control of the child’s behavior.” The subsequent items were used on the survey: “Spanking is a normal part of parenting,” “Sometimes the only way to get a child to behave is with a spank,” “When all is said and done, spanking is harmful for children,” and “Overall, spanking is a bad disciplinary technique.” The first 2 items were reverse scored and an average score for the 4 items was calculated. Responses were rated on a 5-point Likert scale: 1 = strongly agree, 2 = somewhat agree, 3 = neutral, 4 = somewhat disagree, and 5 = strongly disagree. A higher score indicated having a more positive attitude toward CP.

Perceived Injunctive Norms of Colleagues Regarding Corporal Punishment

Next, respondents were asked to tell us their “perception of the opinion of the average primary care pediatrician in your state toward spanking children.” The same 4 ATS items were used with the same 5-point Likert scale. Again, the first 2 items were reverse scored, an average score for the 4 items was calculated, and higher scores indicated more perceived positive attitudes of colleagues toward CP. Because no prior measure of this construct existed, we developed one by modifying the introduction to the ATS scale. A similar modification was originally tested among parents asking about perceptions of attitudes toward CP among close family and friends (α = 0.83) as well as among professionals (α = 0.82).¹³ More recently, this modification was made to test norms among another group of professionals, and a high reliability score was obtained (α = 0.86).²⁰

Beliefs: Positive and Negative Expected Outcomes of Using Corporal Punishment

This measure was modeled after the Outcomes of Physical Punishment Scale, which tested only 7 items in 2 samples: 3 positive outcome items (α = 0.71 and 0.78) and 4 negative outcome items (α = 0.79 and 0.87).²⁴ Additional scale items were added based on newer research linking CP with more negative outcomes, research citing reasons parents tend to use CP, and also in an effort increase reliability scores.^1,25 Respondents were asked: “How often do you think that spanking leads to the child…” followed by 8 positive outcomes and 6 negative outcomes. The “positive outcomes” subscale included: (1) having better behavior in the short term, (2) having better behavior in the long term, (3) being more respectful of parents, (4) learning correct behavior, (5) having a better relationship with the parent, (6) having a decreased chance of delinquency, (7) having a decreased chance of incarceration, and (8) having a better sense of self-control. The “negative outcomes” subscale included: (1) being physically injured, (2) being more aggressive, (3) being physically abused, (4) having poorer cognitive abilities, (5) having poorer mental health, and (6) having poorer physical health. Participants rated each item on a 5-point Likert scale: 1 = never, 2 = seldom, 3 = sometimes, 4 = most of the time, and 5 = always. Scores were averaged so that final scores could range from 1 to 5. A prior study found high reliability for the positive outcomes (α = 0.88) and negative outcomes (α = 0.89) subscales.²⁰ For Figures, findings for “always” and “most of the time” were merged because less than 1% of the sample chose “always” for each item for positive expected outcomes of CP and less than 7% of the sample chose “always” for each item for negative expected outcomes of CP.

History of Corporal Punishment in Childhood

Experiencing CP in childhood is a strong predictor of having positive attitudes toward CP in adulthood.^13,26,27 Therefore, we asked: “While you were growing up, in your first 18 years of life, how often did any of your parents, including step-parents or other caregivers, spank or hit you with a hand or an object for disciplinary purposes?” Options were never, rarely, sometimes, and often.

Analysis

Summary and descriptive statistics were conducted on all measures. Multivariable regression analyses were conducted to see how attitudes toward CP, perceived attitudes of colleagues toward CP, and expected outcomes of CP (positive and negative) were associated with respondent characteristics. All analyses were conducted using STATA.²⁸

RESULTS

Demographic characteristics of the study sample are shown in Table 1. All survey respondents were pediatricians, and most were members of the American Academy of Pediatrics (85.0%). Most had 15 years or more of experience (66.2%), and 12% had completed a post-residency program. Most were white (69.3%), with fewer identified as Asian (14.9%), Hispanic (6.5%), black (3.4%), and other (4.2%). The majority were female (59.1%) and the average age was 51 years (SD, 11.3; range, 19–84). Nearly all identified themselves as primary care clinicians (93.8%) versus specialists and as general pediatricians (92%) versus family practice, internal medicine/pediatrics, or something else. Most respondents worked in a private practice office (55.5%) or in a community-based office or clinic (26.6%), while very few worked in a hospital clinic (11.6%) or other setting (5.5%). Respondents saw an average of 86 (SD, 46.2) children during a typical week in their practice, with a mean number of wellness visits conducted equal to 37.4 (SD, 35.5). The majority (97.1%) reported that children less than 6 years of age represented the largest portion of children seen in their practice. Nearly half (45.1%) reported that they were “rarely” spanked or hit for discipline while they were growing up, whereas 28% were “sometimes” hit, 18.3% were never hit, and only 6.9% were hit “often.” All regions of the country were represented with nearly one-third (31%) from the Northeast, 28% from the South, 23.4% from the West, and 15.3% from the Midwest.

Table 1.

Sample Characteristics (N = 787)

	n	%a
Gender
Female	465	59.1
Male	315	40.0
Race
White	545	69.3
Black	27	3.4
Asian	117	14.9
Hispanic-identified	51	6.5
Other	33	4.2
Region of US
Northeast	245	31.1
Midwest	120	15.2
West	184	23.4
South	222	28.2
Years of experience
<15	259	32.9
15–24	223	28.3
25+	298	37.8
Spanked as child
Never	144	18.3
Rarely	355	45.1
Sometimes	220	28.0
Often	54	6.9
Type of practice
Private practice	437	55.5
Community-based practice	209	26.6
Hospital	91	11.6
Other	43	5.5
Type of pediatrician
Primary care	738	93.8
Specialist	42	5.3
Area of pediatrics
General pediatrics	724	92.0
Other	56	8.0
Member of American Academy of Pediatrics
Yes	665	85.0
No	107	13.6
Age of children seen
>6 year old	764	97.1
<6 year old	6	0.7
	Mean	SD	Range
Age	51.93	11.3	29–84
Average number of children seen weekly	86	46.2	0–500
Average number of wellness visits weekly	37.4	35.5

^aNot oil percents for each variable add up to 100% due to missing values, which never exceeded 2% for any one variable.

We were able to examine the representativeness of our obtained sample based on 4 variables: gender, region, practice size, and patient volume. There were no statistically significant differences between survey respondents and non-respondents on gender nor on patient volume. However, pediatricians from the West were slightly over-represented (20.9% in sample vs 16.3% in non-respondents, 58% response rate) and those from the South were under-represented (28.3% in sample vs 32.5% in non-respondents, 50% response rate). Also, physicians in larger practice settings were under-represented (23.5% in sample vs 29.3% in non-respondents, 47% response rate) compared to those in smaller practice settings (76.5% in sample vs 70.7% in non-respondents, 54% response rate). This latter finding was not surprising, as the mailed surveys likely had a harder time reaching physicians in larger practices.

Most pediatricians who responded to the survey had negative attitudes toward spanking. The majority (78.8%) agreed (strongly or somewhat) that spanking is a bad disciplinary technique overall and that spanking is harmful for children (71.9%), while disagreeing that spanking is sometimes the only way to get a child to behave (77.8%) or that spanking is a normal part of parenting (75.1%). They anticipated slightly higher levels of approval of spanking from their colleagues, but still expected that most would agree that spanking is a bad disciplinary technique overall (71.9%) and that spanking is harmful for children (65.7%); most also thought that their colleagues would not see spanking as being a normal part of parenting (65.4%) or as sometimes being the only way to get a child to behave (64.3%). This latter item produced the largest gap (13.2%) between respondents’ own attitudes toward corporal punishment (CP) and their perceived attitudes of colleagues, whereas the “spanking is harmful item” produced the smallest gap (6.2%).

Pediatricians predicted more negative outcomes from using CP than positive outcomes. As shown in Figure 1, most thought that CP never or seldom resulted in the child having a better relationship with the parent (81.9%), being better behaved in the long-run (79.5%), having a better sense of self-control (79.2%), having a decreased chance of later incarceration (77.3%), having a decreased chance of later delinquency (73.5%), being more respectful of parents (72.5%), and learning correct behavior (66.2%). Just over half (51.7%) believed that CP sometimes results in the child being better behaved in the short-term. Conversely, as shown in Figure 2, most respondents indicated that CP sometimes, most of the time, or always results in a child being more aggressive (86.7%), being physically abused (70.3%), having poorer mental health (68.8%), being physical injured (65.3%), and having poorer physical health (50.3%). Less than half of the sample (45.7%) believed that spanking sometimes, most of the time, or always results in a child having poorer cognitive abilities.

Figure 1.

Perceived positive outcomes of corporal punishment.

Figure 2.

Perceived negative outcomes of corporal punishment for a child.

All 4 of the main constructs we examined were highly correlated (p < 0.001). “Positive attitudes toward CP” was strongly positively correlated with “expected positive outcomes of using CP” (r = 0.66), strongly negatively correlated with “expected negative outcomes of using CP” (r = −0.55), and moderately positively correlated with “perceived positive attitudes of colleagues toward CP” (r = 0.46). “Expected positive outcomes of using CP” was moderately negatively correlated with “expected negative outcomes of using CP” (r = −0.47); and “perceived positive attitudes of colleagues toward CP” had smaller correlations with the former (r = 0.30) and the latter (r = −0.22) variables.

Summary statistics for these 4 variables in Table 2 show that most pediatricians did not have favorable attitudes toward CP as a parenting strategy. Further, they perceived similar unfavorable attitudes toward CP amongst their colleagues, although slightly more positive than their own. Pediatricians expected CP to lead to more harm than good, with most indicating that they “seldom” expected positive outcomes from CP use and “some of the time” expected negative outcomes from CP use.

Table 2.

Descriptive Statistics of Measures Assessing Attitudes, Perceived Norms, and Expected Outcomes Regarding Corporal Punishment (N = 787)

Measures	N	Alphaa	Mean	SD	Strongly Agree/Agree	Neutral	Strongly Disagree/Disagree
1. Positive attitudes toward CPb	783	0.86	1.86	0.97	6.6%	19.5%	74.0%
2. Perceived positive injunctive norms toward CPb	747	0.84	2.07	0.86	4.5%	31.9%	63.7%
Measures	N	Alphaa	Mean	SD	Never/Seldom	Some of the time	Always/Most of the time
3. Expected positive outcomes from using CPc	777	0.91	2.03	0.64	78.4%	19.1%	2.5%
4. Expected negative outcomes from using CPc	760	0.85	2.79	0.63	26.0%	64.4%	9.6%

^aCronboch’s Alpha (α) reliability scores for each of the 4 scales. Scores of 0.8 or above are considered good; 0.9 or above are considered excellent.

^bMeans are calculated based on the following scale: 1 = strongly disagree, 2 = somewhat disagree, 3 = neutral, 4 = somewhat agree, and 5 = strongly agree; higher scores indicate more positive attitudes toward CP. Mean scores of 1 to 2.59 were assigned to “strongly disagree/disagree,” scores of 2.6 to 3.39 were assigned to “neutral,” and scores of 3.4 to 5 were assigned to strongly “agree/agree.”

^cMeans are calculated based on the following scale: 1 = never, 2 = seldom, 3 = some of the time, 4 = most of the time, and 5 = always. Mean scores of 1 to 2.59 were assigned to “never/seldom,” scores of 2.6 to 3.39 were assigned to “some of the time,” and scores of 3.4 to 5 were assigned to “always/most of the time.” CP, corporal punishment.

Table 3 describes variations in pediatricians’ attitudes, perceived norms, and expected outcomes by gender, race, US region, and history of experiencing CP in childhood. Overall, pediatricians that were male, Black, from Southern states, and/or who “sometimes” experienced CP in childhood had more positive attitudes toward CP, perceived more positive attitudes towards CP amongst their colleagues, and expected more positive and fewer negative outcomes from CP than their counterparts (i.e., women, non-black pediatricians, those from non-Southern states, and/or those who “never” or “rarely” experienced CP in childhood, respectively). Pediatricians who experienced CP “very often” in childhood perceived more positive attitudes toward CP amongst their colleagues than those who never or rarely experienced CP. Finally, pediatricians who were female, Asian, from the Northeast, and/or who “never” experienced CP in childhood anticipated more negative outcomes from CP than their respective counterparts (i.e., men, non-Asian pediatricians, those from non-Northeastern states, and/or who experienced CP in childhood).

Table 3.

Multiple Regression to Assess the Association Between Demographic Characteristics and History of Being Spanked as a Child, and Attitudes, Perceived Norms of Colleagues, and Beliefs (Positive and Negative Expected Outcomes) About Corporal Punishment (N = 787)

	Positive Attitudes toward CP				Perceived Positive Injunctive Norms toward CP				Positive Expected Outcomes of Using CP				Negative Expected Outcomes of Using CP
	B	SE B	β		B	SE B	β		B	SE B	β		B	SE B	β
Gender (ref = female)
Male	0.25	0.07	0.12	**	0.08	0.06	0.04		0.17	0.05	0.13	***	−0.18	0.05	−0.14	***
Race (ref = white)
Black	0.74	0.19	0.14	***	0.46	0.18	0.10	**	0.35	0.13	0.10	**	−0.07	0.13	−0.02
Asian	0.08	0.10	0.03		−0.06	0.09	−0.03		−0.06	0.06	−0.03		0.27	0.06	0.15	***
Hispanic	0.22	0.14	0.05		0.13	0.13	0.04		0.08	0.09	0.03		−0.02	0.10	−0.01
Other	0.22	0.17	0.04		−0.17	0.16	−0.04		0.07	0.12	0.02		−0.04	0.12	−0.01
Region of US (ref = Northeast)
Midwest	0.21	0.11	0.08	*	0.36	0.10	0.15	***	0.05	0.07	0.03	***	−0.05	0.07	−0.03
West	0.10	0.09	0.04		0.11	0.08	0.06		0.08	0.06	0.05		−0.15	0.06	−0.10	*
South	0.41	0.09	0.19	***	0.44	0.08	0.23	***	0.27	0.06	0.19		−0.13	0.06	−0.09	*
Years of experience (ref < 14 yr)
15–24	−0.08	0.09	−0.04		0.03	0.08	0.02		−0.05	0.06	−0.04		0.06	0.06	0.04
25+	−0.14	0.08	−0.07		0.02	0.08	0.01		−0.14	0.05	−0.11	**	0.07	0.06	0.06
Spanked as child (ref = rarely)
Never	−0.19	0.09	−0.08	*	−0.04	0.09	−0.02		−0.23	0.06	−0.14	***	0.13	0.06	0.08	*
Sometimes	0.38	0.08	0.18	***	0.18	0.07	0.10	*	0.13	0.05	0.11	**	−0.13	0.05	−0.10	*
Often	0.11	0.14	0.03		0.44	0.13	0.13	***	0.13	0.09	0.05		0.05	0.09	0.02
R2 (variance explained)	0.128				0.098				0.122				0.083

Estimated effects of each variable are represented by: B, linear regression coefficient; SE B, standard error of B; β, standardized regression coefficient. Coefficients >0 mean there is a positive association between the X and the Y variable; coefficients <0 mean there is a negative association. Those that are statistically significant are indicated as follows:

^*p < 0.05

^**p < 0.01

^***p < 0.001.

CP, corporal punishment.

DISCUSSION

Findings from the current study provide an important update to our understanding of pediatricians’ injunctive norms and beliefs about parents’ use of corporal punishment (CP). Roughly 3-quarters of this national sample of pediatricians held negative views toward CP and did not endorse its use. That stands in contrast to the 59% that approved of its use 25 years ago.¹⁶ It also stands in contrast to the 65% to 76% of US adults who believe CP of children is a necessary form of discipline.⁶ This new finding suggests that US pediatricians’ attitudes have changed substantially about this issue over the past couple of decades. Overall, these findings bode well for children’s health.

Our data on pediatricians’ perceived norms regarding CP amongst their colleagues are unique. To our knowledge, no other published study has examined this construct among a national sample of pediatricians in the US. In general, most pediatricians’ in our study perceived that their colleagues also did not approve of CP. However, a sizable amount believed that their colleagues had more favorable views of CP that they did, with the largest discrepancy found for the item “Sometimes the only way to get a child to behave is to spank.” That is, many thought that their colleagues were more likely to endorse this item than they themselves were. This discrepancy might make some pediatricians (who are opposed to CP) less likely to make their opinions on this topic known to their colleagues as some of them may be experiencing some “pluralistic ignorance.”²⁹ By eliminating this pluralistic ignorance (i.e., clarifying that there is an overall injunctive norm of pediatricians against the use of CP), results of this study may increase the willingness of pediatricians opposed to CP to voice their opinions to colleagues as well as parents.

Our findings regarding pediatricians expected outcomes of CP use also are unique. Pediatricians were very knowledgeable about the latest empirical evidence that strongly links CP with increased risk for negative outcomes in children. In particular, they were, correctly, most concerned about CP increasing risk for aggression, being physically abused, and mental health problems.^1,30,31 The fact that these beliefs were strongly positively correlated with unfavorable attitudes toward CP suggests that educating pediatricians even more about this empirical evidence is important.

Importantly, all measured attitudes, perceived norms, and expected outcomes regarding CP varied by some key demographic characteristics. In general, pediatricians who were male, black, from the South or Midwest, or who were “sometimes” (versus never or rarely) spanked as a child were more supportive of CP use than their counterparts. Similar trends were found both for perceived norms regarding CP amongst colleagues as well as for expecting positive outcomes from CP. These findings are not surprising as they reflect the same pattern of increased support for CP that has been reported in the General Social Survey.⁶ This national survey found that US adults who are male, black, or from the South or Midwest were more likely than their counterparts to believe that giving a child a “good hard spanking” is sometimes necessary for good discipline. Also, it is well established in the literature that persons who experienced CP in childhood are more likely to support its use than those who did not.^13,26,27 These findings serve as an important reminder that pediatricians are subject to the same social normative influences and variations as the general population. Any efforts within the profession to change pediatric guidelines or recommendations would benefit from having an awareness of and sensitivity to these group differences.

This study had some important limitations. First, it is not clear how the current findings compare directly with prior surveys on this topic given that different survey measures were used and items about CP were asked in different ways. Similar prior surveys are cited for comparison, but it is not possible to tell if specific contextual attitudes have shifted with time or not. Second, although the response rate was highly respectable, it is possible that non-respondents differ from those who completed the survey and that the results might not be generalizable to all primary care pediatricians. However, our sample is quite similar to the American Academy of Pediatrics (AAP) membership of 2016 regarding gender (64% female) and ethnicity (73% White).³²

This study has important public health implications due to the poor health outcomes associated with CP,¹ the fact that most physicians at some point witness parent-to-child CP in medical settings,¹⁹ and the power that pediatricians have to advise and influence parents on this topic.^13,14,33 Unfortunately, this updated pediatric opinion related to CP may not be reaching most parents. Despite recommendations from the AAP that pediatricians counsel parents about discipline on a routine basis, less than half of parents report receiving education about discipline as part of a well child visit.^34,35 Physician time constraints are likely an important barrier to parents receiving this education. However, brief interventions designed to change parents’ attitudes about CP can be integrated into the clinic visit.^36,37 More studies are needed to determine effective and efficient ways for the well child visit to be used to its fullest potential to educate parents away from using CP as a form of discipline and provide parents with information about alternative forms of discipline that are healthier and less harmful to children than CP.

It is essential that pediatricians make their empirically-based knowledge and opinions on the topic of CP use known to parents as well as to their colleagues, the general public, the media, and policy-makers. As we have seen with other major public health issues such as smoking, the voice of medical professionals is often needed in order to adequately spread burgeoning scientific knowledge and consensus.³⁸ Pediatricians in particular are key messengers of scientific information to parents because they have both credibility and direct access via wellness visits to offer anticipatory guidance relevant to children’s health.