The predictive values of a deliberative and a paternalistic attitude towards two situations of moral conflict: A study among Dutch nurse practitioners and physician assistants.

BACKGROUND: In this study, we examined the predictive values of a moral deliberate and paternalistic attitude on the propensity of yielding to pressure. In these hypothesised positive and negative relationships, we further sought to ascertain whether moral disengagement plays a pivotal role when individuals deviate from ethical standards, rules and regulations when yielding to pressure. AIM(S): This study's primary aim was to assess the predictive value of a moral deliberative and paternalistic attitude for yielding to pressure when physician assistants (PAs) and nurse practitioners (NPs) face moral conflicts.
METHOD: This validation study was cross-sectional and based on a convenience sample of Dutch PAs and NPs. The MSQ-DELIB and MSQ-PATER scales indicate a moral deliberate or paternalistic attitude. These scales were assumed to have a predictive value towards the degree of yielding to pressure by PAs and NPs. Yielding to pressure was measured by two vignettes in which respondents faced a moral conflict (vignette 1: prescribing unindicated antibiotics and vignette 2: discharging a difficult patient from the hospital).
RESULTS: Only moral deliberation was a significant predictor of yielding to pressure. That is, we found a positive effect in vignette 1 (in which the pressure came from the patient). In contrast, we found a negative relationship in vignette 2 (in which pressure went from the working environment). Paternalism did not affect yielding to pressure in either vignette.
CONCLUSION: This study suggests that PAs and NPs having a moral deliberative attitude makes them receptive to pressure exerted by patients to break moral standards. On the other hand, they are more resilient against doing so when this pressure comes from different sources than the patient. Further research is needed to find more conclusive evidence for this differential effect.

INTRODUCTION

In the last decades, patient behaviour has changed [1], most likely because of the obviousness of shared decision‐making. Patients have increasingly become more articulate and have a strong voice in their treatment. This development has many positive aspects. It gives space to the patient's preferences and ideas about treatment within the interplay of patient and healthcare provider besides offering freedom of choice. A downside of this change in patients’ position is that they negotiate for what they think is a superior treatment option instead of a standard treatment that may be sufficient and cost‐effective [2, 3]. This phenomenon seems to be triggered by the easily accessible medical information available on the Internet [4, 5]. The danger in this is that the patient, as a layperson, may think that (s)he is being denied the most optimal care. In such a situation, all healthcare providers can be trapped by the emerging moral conflict. This conflict between options pushes healthcare providers into a position where they need to weigh interests, which results in decisions that are not in line with the (moral) guidelines. The purpose of the current study was to shed more light on the factors that determine whether healthcare providers yield to external pressure in moral conflicts.

We studied this among a specific population, namely physician assistants (PAs) and nurse practitioners (NPs). The PA and NP master degree programmes in the Netherlands respond to an anticipated medical workforce shortage around the early 2000s. In the Netherlands, both the PA and NP programmes are offered through universities of applied sciences and have a length of 30 and 24 months respectively, during which didactics and clinical training interweave throughout the whole curriculum. The reason for including both these professionals concerns their status of having full practice authority (i.e. independent practice). This autonomy accounts for reserved acts that previously belonged to only the realm of medical doctors. Even though there is a difference in the scopes of practice between PAs and NPs, both learn to perform medical history taking, conduct a physical exam, request and interpret additional diagnostics, render differential diagnoses, set a diagnosis, and consequently determine the treatment plan. The legal autonomy also includes the allowance of prescribing drugs. Like doctors, these professional groups can therefore get into moral conflict situations because of these professional responsibilities. Although we know a lot about both professionals’ clinical performance, ample research has been done into the behaviour when a moral conflict arises.

Moral conflict in relation to moral action or yielding to pressure

Whenever healthcare providers and patients interact, disagreement may arise about beliefs, opinions and values that both parties hold [6]. When these different opinions or demands clash normatively, the philosophical literature speaks of a moral conflict and requires an incompatible action [7].

Our study sees moral conflicts as a state once the most rational option (based on medical standards, guidelines and professional ethos) clashes with the opposite choice. Most often, the contrasting option is an emotionally directed one, the desired one of the patient or involved ones [8]. In other words, a moral conflict is a situation where one option prevails over the other. For example, when the PA or NP proposes an evidence‐based option A for a patient, but the patient (or relatives) prefers a non‐evidence‐based option B, the PA or NP finds him‐ or herself in a conflict situation. Based on their professional stance, PAs and NPs intrinsically want to do good for the patients. However, they learned to consider the patient's as well as the relative's choices. This emotional dimension may blur the correctness of the decision and consequently cause the effect of what we introduce as ‘yielding to pressure’ inflicted by the emotionally driven, steadfast, compelling patient. Once the PA or NP yields to pressure(s), he abandons the route of moral action. This reaction of yielding under patient pressure is not a novelty. It is also a known pitfall in the interaction between patients and doctors. In a study by Little et al. [9], the degree of perceived pressure appears to be a significant predictor of whether someone eventually yields under pressure from the patient. For example, one can imagine a situation where the next of kin of a terminally ill patient claims a novel type of chemotherapy to prolong the life of a beloved one, whereas to the clinician's knowledge, this will only severely impact the quality of the short, remaining life [10]. To say, the conflict between wanting to be perceived by the family as a good, involved clinician instead of the professional duty of alleviating a patient's suffering. However, moral conflicts with a more marginal (perceived) impact, such as the moderately ill patient who persists in getting antibiotics without any legitimate indication, viewing that as their right [11] can also be experienced as a moral conflict. In such a situation, the patient's demands conflict with the generic responsibility of PAs and NPs to prevent antibiotic resistance, but also the desire to keep a good understanding with the patient.

In conflicts such as the above, the factors that make a PA and NP more likely to resist yielding to pressure, and make an ethically and medically justified choice for the right course of moral action, are varied. In this paper, we focus on specific attitudes of the PA and NP that may determine this moral action, namely: moral deliberation, paternalism and the propensity to disengage morally.

Moral deliberation and paternalism as predictors of moral action

In an earlier study, we found that PAs and NPs adhere to one of the two types of attitudes when encountering a patient: moral deliberate (MSQ‐DELIB) and paternalistic (MSQ‐PATER) attitude [12]. We defined moral deliberation as a type of medico‐ethical decision‐making act to help patients determine the best health‐related values realised in the clinical situation after considerable deliberation. PAs and NPs with a high propensity towards moral deliberation focus on patient wishes rather than professional norms and values. On the other hand, paternalism entails that clinicians prefer arguments based on rules and regulations in their decision‐making. Decisions are established through the interplay between the clinician's opinion, medical knowledge, experience, colleague's opinions while completely ignoring the patient's will. Paternalistic PAs and NPs will be less interested in engaging with patients.

We assume that during moral conflicts, yielding to pressure would depend on both deliberation and paternalism. When someone has a deliberate moral attitude, there is a high tendency to focus on the patient's wishes and be more sensitive to appeals from the patient or their environment. The result is that the healthcare professional tempts to give in to the pressure at the cost of medical standards, guidelines and professional ethos. We, therefore, hypothesise that:

Moral deliberation has a positive relationship with yielding to pressure.

In contrast, when a healthcare provider has a more paternalistic stance(s), he wants to adhere to the rules and professional standards. Therefore, it is not likely that paternalists will go along with the desires of the patient desire and yield to pressure.

Paternalism has a negative relationship with yielding to pressure.

The dark side of yielding to pressure: moral disengagement

Although moral deliberation contributes to yielding to pressure at the cost of medical standards and guidelines, this may not come without personal costs for the health practitioner. Complying with a patient's request against the moral rules can threaten the healthcare provider's self‐image. For persons to come this far, they need to deal with this somehow. One way to do this is moral disengagement. Moral disengagement defines the process of cognitive reframing of conduct as being morally acceptable without the necessity of changing one's moral standards [13]. There are various ways to reframe immoral acts into moral ones: downplaying the harmful consequences, using euphemisms to make it sound less bad, or shifting the responsibility for the behaviour to someone else [14]. These moral disengagement ways make it easier for people to deviate from moral standards, rules and regulations without feeling guilty [15].

Concerning yielding to pressure, one can imagine that every individual has an internal standard that prohibits deviation from moral action. After all, moral action is dictated by rules and regulations in addition to professional ethos, or rather, the inner feeling of the way it ought to be. Nonetheless, when the force is too strong to resist, and someone yields to the pressure, moral disengagement mechanisms may facilitate the PA and NP to construe a new, convenient ‘truth’.

For example, when pressured into prescribing antibiotics without an indication, PAs and NPs may tell themselves that prescribing this desired medication unindicated is a minor issue compared to the action of other colleagues who violate opioid regulations. They may also reveal that patients are illegally buying antibiotics online already, so they may better prescribe them when they insist. Such ‘excuses’ that a health professional can tell him‐ or herself can render the ethical misconduct as unrelated to the own moral standards against deviating from medical rules, regulations or even professional ethos. This thought helps the healthcare provider prescribe to the belief that nothing is wrong.

Considering the above example, it is clear that the interrelated mechanisms of moral disengagement facilitate unethical behaviour. On this basis, we expect that professionals scoring high in moral deliberation, that is, those who tend to go along with patient's demands, can only do so if they can justify the morally questionable behaviour for themselves. So, only those who are also prone to moral disengagement will go along with the patient's demands. In other words, moral deliberation will only increase yielding to pressure when moral disengagement is high and not when it is low. For this, we hypothesise the following:

Moral deliberation and moral disengagement will interact to predict yielding to pressure in a way that moral deliberation will positively predict yielding to pressure when moral disengagement is high rather than low.

We have no reason to expect that moral disengagement moderates the influence of paternalism on yielding to pressure. After all, we expect paternalists to be unreceptive to pressure. Therefore, the paternalists may not be inclined to deviate from their self‐convinced course of action and do not need to use moral disengagement mechanisms.

In sum, in our study, we aim to assess the influence of a moral deliberative and paternalistic attitude on yielding to pressure when PAs and NPs are confronted with moral conflicts and consider whether moral disengagement plays a moderating role in this. The relevance of our study lies in the fact that immoral behaviour in medical situations is undesirable. Understanding factors that may trigger or explain this behaviour can help reduce/prevent this unwanted behaviour. Preferably, attention is paid to this during the training period of healthcare professionals already.

METHOD

Study design, participants and data collection

In this cross‐sectional study, five PA degree programmes and one NP degree programme served as sources for approaching alumni. As per the European General Data Protection Regulation, the researchers had no permission to use the databases of the programs to retrieve the email addresses of alumni. For this reason, we sent letters explaining the study to the programme administrators, who mailed them to their respective PA and NP alumni. The letter contained a hyperlink to a private web‐based system (name). If willing to participate in the survey, the alumni activated the hyperlink and provided their email contact details. Of the 896 alumni (470 NPs and 426 PAs) the programme administrators sent letters to, 294 (176 PAs and 118 NPs) provided their e‐mail addresses. We sent an access key to the web‐based study questionnaires to these alumni who provided their email addresses. At the end of the online survey period (January–March 2015), 155 respondents had completed all of the questionnaires (response rate of 52.7%). We could not test for selection bias, as no information was available about the alumni who did not participate. To prevent missing data, we designed all the survey questions in the forced‐choice format.

The dataset used in the current study was the same as the one in previous studies by Kuilman and colleagues [12, 16, 17]. Variables from that pool were used in the present study but were used to address different hypotheses.

Measurements

Sociodemographic characteristics

The following background characteristics were collected to conduct tests for the comparability of the NP and PA samples: gender, age, religious beliefs and political affiliation. Respondents also characterised their working environments as (a) ‘hospital’; (b) ‘general practice’; (c) ‘mental healthcare’; (d) ‘care for people with mental disabilities’; or (e) ‘other’.

Indicator of Yielding to pressure

In this study, we used two vignettes as indicators of ‘yielding to pressure’. These vignettes, as exhibited in Appendix 1, are regarded as two separate indicators as they tap two distinct dimensions of moral conflict that occurs: (1) during provider–patient interaction (vignette 1) and (2) during provider–colleagues interaction within the working environment (vignette 2). They both indicate degrees to which PAs and NPs yield to pressure during a moral conflict. On a scale from 0 to 100, respondents indicated how likely they are to act in the following ways: (a) prescribe antibiotics without a medical indication to a demanding patient (vignette 1) and (b) discharge a schizophrenic patient from the hospital with oral antibiotics, pressured by the demanding nursing staff to restore calm and order in the nursing ward (vignette 2). A higher score on both vignettes indicated a higher likelihood of yielding to pressure at the cost of adherence to rules and regulations.

Indicators of moral deliberation and paternalism

In an earlier study, we validated the two scales MSQ‐DELIB and MSQ‐PATER, as measures of moral deliberate and paternalistic attitude respectively. Both scales have a good internal consistency, as indicated by Cronbach's alpha of 0.70 (17). The 4‐item MSQ‐DELIB contains items like: ‘As a PA/NP, I must always know how individual patients in my ward should be respectfully approached’ and ‘What is most important in my clinical practice is my relationship with the patients’. The 7‐item MSQ‐PATER scale contains items like (a) ‘I always base my actions on the medical knowledge of what is the best treatment, even if the patient protests’ and (b) ‘When I need to make a decision contrary to the will of a patient, I do so accordingly to my opinion about what is good care’.

Moral disengagement scale

To measure propensity to disengage morally, we modified the moral disengagement scale of Bandura et al. to fit the perspective of general healthcare [14]. For example, item number 32 in Bandura's scale: ‘Children are not at fault for misbehaving if their parents force them too much’ was replaced by: ‘Medical professionals cannot be held accountable for their mistakes when the government puts them under heavy pressure’. We invited the respondents to answer 32 statements on a Likert‐type scale (1 = completely disagree to 5 = completely agree). The scale's Cronbach's alpha of 0.85 in our study was consistent with the findings by Bandura et al. and indicated that translation and adaptation did not affect the internal consistency of the scale. We employed an algorithm to calculate an overall scale score by subtracting the minimum scale score from the raw scale score, dividing this by the scale score's range, multiplied by 100, resulting in scores ranging from 0% to 100%. A higher score indicated a higher propensity to disengage morally.

Statistical analysis

We did a multiple stepwise regression analysis and calculated cross‐products for the interaction terms to test our hypotheses, all based on the variables transformed towards normality [18]. Step 1 included the variables age and gender, followed by either moral deliberation or paternalism and moral disengagement in step 2. During step 3, we added the interaction terms to the model, each linked to the independent variable.

Ethical considerations

According to the Dutch Central Committee on Research Involving Human Subjects (www.ccmo.nl), institutional review board approval was not warranted for this type of survey, requiring only the voluntary participation of professionals. An information letter sent to all respondents notified them of (a) purpose of the study, (b) the voluntary nature of participation and (c) their right to stop participating in the study at any time. The respondents got informed that their answers would be completely anonymous, and the information collected would not be used for any purpose other than the study. Furthermore, the letter mentioned the expected average time needed to complete the questionnaires (45 min). This study adheres to the tenets of the Declaration of Helsinki [19]. Only the first author (LK) had access to the encrypted data. The ‘Strengthening the Reporting of Observational Studies in Epidemiology’ (STROBE) checklist served as a guideline for reporting observational research.

RESULTS

Sociodemographic characteristics

For an overview of the sociodemographic characteristics of the respondents, see Table 1. The average age of the respondents was 45.2 years (±9.1), and the majority (70.3%) of the respondents were women. Less than half (46.5%) of the 155 respondents reported being religious and 13.5% indicated a tendency to vote for a conservative political party. Most of the respondents (72.9%) work in hospitals, with a smaller share (14%) working in family medicine (general practice) and the rest working either in mental healthcare (5.8%), care for people with mental disabilities (1.3%) or elsewhere (12.9%). Table 2 presents the sociodemographic variables and correlations between them.

TABLE 1

Socio‐demographic characteristics of participants stratified to physician assistants and nurse practitioners

Sociodemographic characteristics		Physician assistant N = 88	Nurse practitioner N = 67	Total N = 155	(p‐value)
Age mean (SD)		42.5 (8.4)	48,8 (8.7)	45.2 (9.1)	<0.001a
Gender	Female N (%)	56 (63.6)	53 (79.1)	109 (70.3%)	0.05
	Male N (%)	32 (36.4)	14 (20.9)	46 (29.7%)
Religion	Not religious	48 (54.5)	35 (52.3)	83 (53.5%)	0.54
	No denomination but spiritual	3 (3.4)	4 (4.5)	7 (4.5%)
	Christian	35 (39.8)	25 (37.3)	60 (38.7%)
	Muslim	1 (1.1)	0	1 (0.7%)
	Other religions	1	3 (4.5)	4 (2.6%)
Working environment	Hospital N (%)	64 (72.7)	49 (73.1%)	113 (72.9%)	0.58
	General practice N (%)	13 (14.8)	7 (10.5%)	20 (12.9%)
	Mental healthcare N (%)	3 (3.4)	6 (9%)	9 (5.8%)
	Disability care N (%)	1 (1.1)	1 (1.5%)	2 (1.3%)
	Other N (%)	7 (8)	4 (5.9%)	11 (7.1%)
Political orientation	Conservative N (%)	15 (17)	6 (9%)	21 (13.5%)	0.14
	Liberal N (%)	73 (83)	61 (91%)	134 (86.5%)

Independent Sample's T‐test.

TABLE 2

Average scores and correlations across the scales themselves and with sociodemographic parameters

Sociodemographics	Variable number	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]	[9]
Age	[1]
Gender	[2]	0.041
Religion	[3]	0.003	−0.039
Political orientation	[4]	0.167a	−0.032	−0.160a
Working environment	[5]	−0.008	−0.001	−0.033	0.148

Correlation is significant at the 0.05 level (two‐tailed).

Correlation is significant at the 0.01 level (two‐tailed).

Multivariable regression analysis

Upfront of all interpretations of the outcomes, we first assessed for multicollinearity, as that might be a potential threat in a cross‐sectional data collection method. For this, we used two essential parameters, namely ‘Tolerance’ and ‘Variance Inflation Factor’ (VIF). Considering all the Tolerances being well above 0.1 and all VIFs far below 10 (see Tables 3 and 4), we excluded the presence of multicollinearity that could have affected the outcomes [20].

TABLE 3

Multiple regression analysis with moral deliberation as the independent variable

Model		Vignette 1 ‘Unindicated antibiotics’				Vignette 2 ‘Schizophrenic patient’
		Beta	Sig.	Collinearity statistics		Beta	Sig.	Collinearity statistics
				Tolerance	VIF			Tolerance	VIF
1	(Constant)		0.005				0.001
	Age	−0.029	0.721	0.998	1.002	0.085	0.294	0.998	1.002
	Gender	−0.181	0.024	0.998	1.002	−0.027	0.738	0.998	1.002
2	(Constant)		0.006				0.002
	Age	−0.013	0.870	0.976	1.024	0.109	0.172	0.976	1.024
	Gender	−0.211	0.007	0.983	1.018	−0.027	0.731	0.983	1.018
	Moral deliberation	0.233	0.003	0.969	1.032	−0.271	0.001	0.969	1.032
	Moral disengagement	0.219	0.006	0.942	1.062	0.040	0.619	0.942	1.062
3	(Constant)		0.005				0.001
	Age	−0.022	0.784	0.953	1.049	0.088	0.268	0.953	1.049
	Gender	−0.215	0.006	0.978	1.022	−0.036	0.648	0.978	1.022
	Moral deliberation	0.243	0.003	0.942	1.061	−0.249	0.002	0.942	1.061
	Moral disengagement	0.218	0.007	0.942	1.062	0.038	0.636	0.942	1.062
	DELIB*MDSa	0.057	0.468	0.946	1.057	0.131	0.102	0.946	1.057

Cross‐product of Moral deliberation × moral disengagement

TABLE 4

Multiple regression analysis with paternalism as the independent variable.

Model		Vignette 1 ‘Unindicated antibiotics’				Vignette 2 ‘Schizophrenic patient’
		Beta	Sig.	Collinearity statistics		Beta	Sig.	Collinearity statistics
				Tolerance	VIF			Tolerance	VIF
1	(Constant)		0.005				0.001
	Age	−0.029	0.721	0.998	1.002	0.085	0.294	0.998	1.002
	Gender	−0.181	0.024	0.998	1.002	−0.027	0.738	0.998	1.002
2	(Constant)		0.012				0.003
	Age	−0.005	0.950	0.929	1.077	0.101	0.232	0.929	1.077
	Gender	−0.201	0.014	0.946	1.057	−0.040	0.629	0.946	1.057
	Paternalism	−0.010	0.905	0.883	1.132	0.017	0.841	0.883	1.132
	Moral disengagement	0.182	0.026	0.946	1.058	0.082	0.326	0.946	1.058
3	(Constant)		0.013				0.003
	Age	−0.006	0.944	0.924	1.082	0.097	0.253	0.924	1.082
	Gender	−0.201	0.015	0.945	1.058	−0.042	0.620	0.945	1.058
	Paternalism	−0.008	0.926	0.839	1.192	0.030	0.734	0.839	1.192
	Moral disengagement	0.181	0.028	0.936	1.068	0.077	0.362	0.936	1.068
	PATER*MDSa	0.009	0.916	0.933	1.072	0.055	0.517	0.933	1.072

Cross‐product of paternalism × moral disengagement.

As can be seen in Table 2, both age and gender correlated with Paternalism. For this reason, we included these variables in the multiple regression in Step 1. In explaining the yielding to pressure in vignette 1 (unnecessary prescription of antibiotics), gender alone remained a significant predictor of yielding to pressure (see Tables 3 and 4). Interpretation of this outcome learns that male (coded as ‘1’) providers in this study are less prone to yield to pressure. This effect, however, was not the case for yielding to pressure in vignette 2.

Predictors of Yielding to pressure

Regarding hypotheses 1 and 2, we assumed that both moral deliberation and paternalism would regress positively and negatively respectively on the propensity of yielding to pressure in vignettes 1 and 2. However, only hypothesis 1 could be partly affirmed (see Table 3) to yielding to pressure in vignette 1. That is, even though moral deliberation behaves as a predictor for both vignettes, for vignette 1 there is a positive relationship (β = 0.244, t = 3.062, p = 0.003) and for vignette 2, moral deliberation turns out to be a negative statistically significant (β = −0.252, t = −3.126, p = 0.002) predictor.

Furthermore, we also had to reject hypothesis 3. Moral disengagement did neither moderate (cross‐product: DELIB*MDS) the relationship between moral deliberation and yielding to pressure in vignettes 1 and 2, nor did it moderate (cross‐product: PATER*MDS) the relationship between paternalism and the propensity of yielding to pressure.

DISCUSSION

The study's primary aim was to assess the predictive value of a morally deliberative attitude and a paternalistic attitude on yielding to pressure when PAs and NPs face a moral conflict. We expected that the deliberate moral attitude would increase (H1), and the paternalistic attitude would decrease (H2), yielding to pressure. Also, we expected the cognitive process of moral disengagement to have a strengthening effect on the relationship between moral deliberative attitude and the propensity of yielding to pressure (H3).

The data gave partial support for hypothesis 1 as moral deliberation positively predicted yielding to pressure in the antibiotic scenario. However, it negatively predicted yielding to pressure in the schizophrenic patient scenario. Both these effects were not moderated by a propensity to disengage morally, rejecting hypothesis 3. Paternalism did not affect yielding to pressure in either vignette, therefore rejecting hypothesis 2.

Although moral deliberation regresses positively on yielding to pressure in vignette 1 (unindicated antibiotics), it is remarkable that it negatively regresses to yielding to pressure in vignette 2 (schizophrenic patients). A possible interpretation of this may lie in the different sources of pressure in both scenarios. In vignette 1, it is the patient himself who exerts pressure on the PA or NP. In that sense, the patient is the subject of the story in vignette 1, whereas, in vignette 2, the nursing staff puts pressure on the clinician to dismiss the patient to restore calm and order. Since a PA or NP with a high degree of moral deliberation attitude is entirely focused on the patient, it makes sense that (s)he is more likely to yield to pressure when a patient exerts pressure (e.g. in vignette 1). In contrast, (s)he is less likely to yield to pressure when this pressure is exerted by someone who chooses side against the patient (e.g. in vignette 2). Whether the source of the pressure (patient, colleagues, administration or the patients’ family) influences the direction of moral deliberation is an exciting avenue for further research.

Furthermore, we expected a negative relationship between paternalism and yielding to pressure since individuals with a paternalistic stance will adhere to the rules and their professional standards at all times and thus would be less likely to yield to pressure to deviate from these rules and standards. However, the results show no relation between paternalism and yielding to pressure. Thus, at this moment, there is no credible evidence to support our hypothesis (H2). Looking at the results, we also see no reason to expect that a significant relationship elicits when retesting the hypothesis among a larger sample. As measured by our paternalism scale, adherence to one's decision, rules and guidelines are unrelated to ‘yield to pressure’. It may be more fruitful in further research to focus on other personality traits that measure persistence more directly and are not necessarily related to the specific medical context.

Theoretical contribution

The most important theoretical contribution of the current paper lies in the finding that, among health practitioners, a high moral deliberation may increase the inclination to yield to pressure exerted by patients. The fact that people fall into morally questionable behaviour by others is a well‐established finding in the domain of psychology and business literature [21, 22, 23, 24]. In that sense, our finding corroborates this literature. Nevertheless, within the context of the PA and NP professional, this has not been highlighted before. Also, previous literature suggests that the tendency to be susceptible to social influences to make (im)moral choices is less robust for people in a powerful position [25]. The explanation for this is that people in high influential roles pay more attention to their thoughts and feelings while people in low influential roles pay more attention to contextual stimuli. Based on this, one would expect that PAs/NPs would be less likely to be influenced by their patients than vice versa. However, our findings suggest that having a deliberate moral mindset can ‘open up’ professionals to focus more on their patients. Hence, our data indicate that the influence of power may be mitigated by moral deliberation, at least in the context of an ethical dilemma in a health context.

Concerning the effects of moral deliberation, our study contributes to further insights into what factors make people more or less likely to yield to pressure. The results of vignette 2 do highlight that moral deliberation can function as a double‐edged sword. It is, of course, a laudable thing to focus on the patients’ interests and needs. However, such focus may also make health professionals vulnerable to influences exerted by the patients that lead them to behave against the moral guidelines, possibly at the disadvantage of other parties or society at large.

Strengths and limitations

One strength of this study is the representative sample being used in terms of gender and age, reflecting the demographics of both the NP and PA workforces in the Netherlands [26]. For this reason, the results of moral deliberation being a predictor of yielding to pressure when occurring in a direct patient–PA/NP interaction can be generalised to the NP and PA. This finding could apply to professionals with comparable independent treatment relationships (e.g. medical doctors, physical therapists, speech therapists or dental hygienists).

In methodological terms, another strength of our study is that we a priori determined the required sample size (n = 68) for multivariable regression analysis using interaction terms, which was well above the factual sample size of 155 respondents [27]. Besides, despite the cross‐sectional nature of the data, Harman's single‐factor analyses indicated that single factors for the different models ranged from 15.0 to 26.4% of the total variance. Given the maximum threshold of 50%, common method variance had little to no effect on the conclusions drawn [28]. Last but not least, both the Tolerance and the VIF used as collinearity diagnostics were well above and below the acceptable thresholds respectively. These findings enabled us to rule out the possible phenomenon of multicollinearity impacting our outcomes [20].

Our study is also subject to several limitations. Even though the correlations between several study variables were statistically significant, their explained variances were relatively low. Therefore, it should be clear that many other factors not included in this study could explain or influence yielding to pressure. Primarily because of the low explained variances, future research needs to explore other factors that could explain the concept of yielding to pressure.

IMPLICATIONS

Our study suggests that a moral deliberate attitude induces a higher risk of yielding to pressure exerted by a patient (vignette 1). In comparison, it causes a lower chance of yielding to pressure exerted by other people in the immediate work environment (vignette 2). Although further research is needed to test the influence of the source of pressure, our findings have implications for how PAs and NPs, and other healthcare professionals are trained. More specifically, habituation of healthcare and nursing students may increase during simulation‐education with scenarios that incorporate aspects of pressure, such as the demanding, aggressive patient. While in training, attention must be paid to dealing with pressure from patients, especially the individuals who have an increased tendency of patient orientation. Also, students should acquire skills and techniques on how to remain patient‐oriented and, at the same time, not yield to pressure is an important aspect. Furthermore, students need to ultimately learn how to stick to moral choices without being led by emotions [29].

CONCLUSION

This study suggests that yielding to pressure is influenced by moral deliberation and not by paternalism. More specifically, it indicates that healthcare professionals with a high degree of moral deliberation are more prone to yield to pressure exerted by a patient and less prone to yield to other types of pressures that seem to go against the patient's interest. Further research is needed to reach more definite conclusions.

CONFLICT OF INTEREST

The authors have no conflicts of interest to disclose.

AUTHOR CONTRIBUTIONS

Luppo Kuilman: study conception/design; acquisition (and storage) of data; analysis and interpretation of data; writing and critical revision of the manuscript; Gerard J. Jansen: study conception/design; drafting manuscript; critical revision; Laetitia B. Mulder: study conception/design of theoretical model; drafting of the manuscript; critical revision; Petrie F. Roodbol: study conception/design; critical revision.

Indicator	Vignette
Yielding to pressure	Coughing for three weeks, ‘I want antibiotics now!’ (vignette 1) You have been working as a [physician assistant/nurse practitioner] at a general practice in Northeast Groningen for several years, and you are now a familiar face, even with patients. On a Friday afternoon at 4.50 pm, just before the consultation hour has ended, Mr. Wolderman, a well‐known tenor, reports to the desk, and with a loud voice, he wants an appointment immediately. He says that he has been suffering from a persistent dry cough for over one week and is demanding antibiotics just before the weekend. The medical history does not report alarm symptoms, the physical examination does not indicate an infection, there is no fever, and the CRP rapid test shows <10 mg/L. In short, you have no indication to prescribe antibiotics. The patient is incensed and still demands a cure in a verbally aggressive manner. You explain that in accordance with the guideline of the Nederlandse Huisartsen Genootschap M78, ‘Acute coughing’, there is absolutely no indication to prescribe antibiotics. Mr. Wolderman kindles in anger because he has a solo part in the Matthew Passion in the Oosterpoort in Groningen.
	The stinky patient! (vignette 2) As a [physician assistant/nurse practitioner], you are the first medical point of contact for all matters that arise daily in the lung medicine nursing department. Last night, a 54‐year‐old homeless man with schizophrenia in poor condition was admitted after a major exacerbation. A day after the admission, a medical assistant reports that the patient spreads an intolerable, pungent stench. As a homeless person, he has been wearing the same clothes for 8 months, 24 hours a day. The patient reacts violently to the offer to wash the clothes because he says: without this ‘magical robe’, I am defenceless against evil. The situation in the room is unsustainable, he doesn't want to shower and no other clothes and his roommates want him to sleep separately. The tension mounts when it appears that a terminal lung cancer patient occupies the only single room. To make matters worse, all surrounding hospitals do not have single rooms available. Your hospital's psychiatry department is prepared to have the patient continue treatment there until he is well again. The patient absolutely does not want this and knows that he cannot be forced to be admitted there. The tensions evoked in this conflicting situation make the patient speak louder and louder in his head. This increases the fear of his roommates even more. You are considering sending him out with enough antibiotics. For the patient, this means back in his homeless life. There is a significant risk that the antibiotics will no longer be taken, and the course of treatment will not be completed.