decision making- week 4

ORIGINAL RESEARCH

Nurse Practitioners’ Versus Physicians’ Diagnostic Reasoning Style and Use of Maxims: A Comparative Study Alison M. Pirret, PhD, NP

ABSTRACT The study used an intuitive/analytic reasoning instrument and maxims questionnaire to compare 1) the diagnostic reasoning style of 30 nurse practitioners (NPs) and 16 resident doctors and 2) its influence on their diagnostic reasoning abilities of a complex case. The results showed NPs incorporated more system I (intuitive) processes when compared with residents; however, both groups identified with certain maxims. Diagnostic reasoning style was not related to participants’ diagnostic reasoning abilities, indicating they triggered system II (analytic) processes when required. Although system I processes are essential, clinicians need to be aware of the value and pitfalls associated with them.

Keywords: diagnostic reasoning style, maxims, nurse practitioner � 2016 Elsevier Inc. All rights reserved.

he first New Zealand (NZ) nurse practi- tioner (NP) was registered in 2002, with

Tthe 100th NP being registered in early

2012.1 In NZ, the title NP is legally protected and can only be used by nurses with a master’s degree who have passed rigorous assessment processes.2

Legislation in NZ allows NPs to practice independently without physician supervision.2

Research shows NPs and resident doctors have similar patient outcomes and diagnostic reasoning abilities.2 Diagnostic reasoning requires clinicians to collect relevant assessment data, retrieve memorized knowledge, and integrate data in the working memory. Because of limited capacity in the working memory, this process can overstretch the cognitive resources and create congitive overload, which risks diagnostic error.3 Singh et al4 suggest 12 million United States adults are affected by diagnostic error every year. Because diagnostic reasoning style impacts on diagnostic accuracy,5 it is worthy of further exploration.

This study compared NPs’ and resident doctors’ diagnostic reasoning style and use of maxims to guide their diagnostic reasoning. It answered 3 questions: 1) How does NP diagnostic reasoning style compare with that of residents? 2) How do maxims used by NPs in their diagnostic reasoning compare with

www.npjournal.org

those used by residents? and 3) Are NPs’ and resi- dents’ diagnostic reasoning ability scores described by Pirret et al2 influenced by their diagnostic reasoning style and use of maxims in everyday practice? The first 2 questions were based on the assumption that as NZ NPs were expected to have more years of experience than residents, they were more likely to use system I processes in their diagnostic reasoning.

BACKGROUND Dual process theory identifies diagnostic reasoning uses system I (intuitive) and system II (analytic) processes; the degree to which each is used is dependent on the clinical situation.5,6 System I processes are fast and are used automatically when clinicians are involved in familiar case presentations. They use cognitive shortcuts or rules of thumb, commonly termed heuristics, to reduce the cognitive load and simplify the diagnostic reasoning process.5,7

These heuristics based on experience, patient characteristics, and the context in which the patient presents enable clinicians to reach a diagnosis without proceeding with the time- expensive process of exploring unlikely diagnoses.5

If the patient presentation is not initially recog- nized, time does not permit, or the clinician is

The Journal for Nurse Practitioners - JNP 381

uncertain, the slower, logical, and deliberate system II processes are used.7 Both system I and II processes need to fail for diagnostic error to occur, system I for making an error and system II for not detecting and correcting it.8 An example of this is when contextual factors, such as clinician overconfidence or fatigue, are combined with a complex case presentation with features that reflect multiple diagnoses,9 such as those described in Box 1. Contextual factors triggered by system I processes create the error, whereas failure to take a complete patient history or perform the appropriate physical examination leads to system II failure. Failure of both the system I and II process leads to diagnostic error and an inappropriate management plan.

Different types of diagnostic reasoning use either system I or II processes. Intuition and pattern recognition use system I processes, whereas the hypothetico-deductive model and clinical guidelines use system II processes. Intuition is based on past experiences, is generated without mental effort, and is commonly described as a gut feeling.6 Pattern recognition is when the clinician makes a diagnosis based on a few pieces of critical information gained from the clinical context and memorized signs and symptoms5; this process allows an almost instantaneous realization that the patient’s presentation resembles memories of previous cases.

Maxims also serve as a heuristic and are thought to aid in memory. They are succinct sayings devel- oped by experienced clinicians.10 Each maxim is case specific and not suitable to be applied to all patients.10

Box 1. Example of

A 61-year-old noneEnglish-speaking woman presents w

posterior chest pain, productive cough with purulent sp

pain. She is accompanied by her daughter, who is able

3 days ago and was prescribed an antibiotic but feels s

managed type II diabetes, hypertension, hyperlipidemi

takes acetaminophen and a nonsteroidal anti-inflamma

weight loss of 8 kg over the past 6 weeks. She is a sm

She is hypertensive with a slightly raised respiratory ra

has dullness over the right base on percussion. Right-s

coughing. Abdominal palpation reveals right upper qu

The Journal for Nurse Practitioners - JNP382

The hypothetic-deductive model is an approach predominantly used by novice clinicians. It uses inductive and deductive reasoning to guide clinicians to the most correct diagnoses.7 Clinical guidelines are used to interpret and treat certain conditions and are thought to be useful in improving the performance of novice clinicians.11

NP Diagnostic Reasoning Styles Most research exploring NP diagnostic reasoning style was published 10 to 20 years ago.12 These studies identified NPs used system I and II processes including intuition, pattern recognition, maxims, and the hypothetic-deductive model.12,13 Intuition was used to alert NPs to issues, which was then followed by a search for more objective data to support their concerns.13 The maxims NPs used included “real disease declares itself, follow-up everything, and common things occur commonly.”13

Medical Doctor Diagnostic Reasoning Styles System I and II processes are incorporated into medical doctor (MD) diagnostic reasoning, but experience determines how they are used. When using pattern recognition, novice doctors use familiar and irrelevant factors to support diagnoses, such as name, occupation, age, and similar situa- tions.14 This is in contrast to experts who support diagnoses with memorized signs and symptoms learned from experience.15

Although MDs use the hypothetico-deductive model, it is only used by experts when analyzing complex or unfamiliar cases.16 Medicine is beginning

a Complex Case

ith a 1-week history of generalized malaise, fevers,

utum, increased shortness of breath, and abdominal

to translate. She visited her general practitioner

he is getting worse. She has a history of poorly

a, and osteoarthritis of her right knee for which she

tory drug. She is obese but has had an unintentional

oker with a 40epack year history.

te. She has right basal crackles on auscultation and

ided chest pain is present on inspiration and

adrant tenderness.

Volume 12, Issue 6, June 2016

to value intuitionwith suggestions thatMDs should be trained to develop confidence in their gut feelings.11

METHODS The study used a comparative research design. This aspect of the analysis used an intuitive/analytic reasoning instrument and a maxims questionnaire to compare NPs’ and residents’ diagnostic reasoning style and the maxims they used to guide their diag- nostic reasoning.

Intuitive/Analytic Reasoning Instrument The intuitive/analytic reasoning instrument was initially designed to measure registered nurse (RN) decision making. The instrument was validated using 1,460 RNs from multiple specialties and countries.17

The instrument uses a 5-point Likert-type scale to measure 56 items, 14 items for each decision-making stage including data collection, data processing, problem identification, and action planning. Of the 56 items, half measure analytic processes, and half measure intuitive processes. Analysis showed there were positive correlations between the items measuring analytic (P ¼ .01-P � .001) and intuitive (P � .001) processes. The total summed scores gained from the recoded 56 items determine 4 types of decision-making styles. Scores < 160 indicate an analytic style, scores � 160 to � 170 indicate an analytic-intuitive or intuitive-analytic style, and scores > 170 indicate an intuitive style.17

Because the instrument was designed to measure RNs’ decision making, for this analysis, the instrument’s wording was reviewed and altered to reflect both NP and resident diagnostic reasoning language. The wording changes did not alter the instrument’s intent, and the instrument’s author approved the changes. Thus, for this analysis, the instrument contained items thatmeasured participants’ use of system I or II processes in their daily practice at each stage of the diagnostic reasoning process (ie, data collection, identifying diagnoses and problems, plan- ning care, and implementing the action plan).

Maxim Questionnaire The maxims questionnaire used a 5-point Likert-type scale to explore the use of 13 maxims in participants’ daily practice. The questionnaire included maxims

www.npjournal.org

identified in the literature review and identified by Bernstein.18

Participants Because this analysis was part of a wider study that used the qualitative think aloud technique,2 statistical advice determined 30 NPs and 30 residents would keep the sample size manageable and show group differences if they were present. Purposeful sampling recruited 30 NPs and 16 residents working in NZ. This left the study underpowered and at risk of a type II error, meaning there was a risk of the research identifying no statistically significant difference between the 2 groups when there was one. Failure to recruit the desired number of residents may have been related to medical resistance to introducing NPs at the time.19

Inclusion criteria required recruited residents to have more than 2 years of experience after their graduation. Residents were used as the comparative group because they often manage complex cases and seek guidance from medical specialists as required. Because NPs practice independently but collaborate with medical specialists when required, they use similar referral/collaboration lines to resi- dents; hence, they may have similar diagnostic reasoning. Because the analysis was part of a larger study that also assessed participants’ diagnostic reasoning abilities of a complex case using a case scenario, participants unlikely to be exposed to the case were excluded. This included those working in child and mental health and specialist areas, such as ophthalmology and organ transplant. No randomization was performed because only 59 NPs at the time of recruitment were eligible for the study. Being able to recruit over 51% of eligible NPs meant the sample was likely representative of NPs working in general adult areas of practice.

Ethical Considerations The Massey University Human Ethics Committee approved the study. All participants provided writ- ten consent.

Data Collection/Analysis Data were collected between February 2011 and March 2012 (inclusive). Each participant accessed

The Journal for Nurse Practitioners - JNP 383

the instrument and questionnaire via a Survey- Monkey (www.surveymonkey.com) web link at a computer and time convenient to them.

Collected data were imported from Survey- Monkey into the Statistical SPSS version 1920 within the instrument enabled calculation of each participant’s total score. These scores indicated diagnostic reasoning style.

The data pertaining to maxims used frequently by NPs were transformed to determine maxims participants mostly identified with in their daily practice. Computing of participants’ data enabled these maxims to be calculated into a single score variable and group comparisons made.

The diagnostic reasoning style and maxim scores were then correlated to participants’ previously reported diagnostic reasoning ability scores related to a complex case.2 This assessed the effect participants’ everyday diagnostic reasoning style and maxims had on their diagnostic reasoning abilities. The alpha level for statistical significance was set at 0.05.

Validity and Reliability An expert panel consisting of a professor of general practice, an associate professor of rheumatology, and an NP reviewed the instrument’s wording to ensure the wording reflected diagnostic reasoning language. The instrument’s reliability was assessed using Cronbach alpha (Cronbach a ¼ 0.89). The maxims were assessed by the expert panel as being suitable to include in the questionnaire. Content validity was improved using purposeful sampling to ensure the NP and resident groups reflected similar specialty areas. Reliability was addressed by ensuring the methods used to assess participants were clearly outlined. The questionnaires used in the research are available online13 to allow the study to be reproduced by others researching similar populations.

RESULTS The data of 30 NPs and 16 residents were analyzed. They worked in multiple specialty areas including general practice, gerontology, cardiology, emer- gency, and palliative and respiratory care. The largest numbers of participants in both the NP and resident groups worked in general practice, and the smallest numbers worked in palliative care.

The Journal for Nurse Practitioners - JNP384

PARTICIPANT DEMOGRAPHICS NPs Twenty-seven female and 3 male NPs participated in the study. They had a mean (M) of 2.2 years (standard deviation [SD] ¼ 1.6) of NZ NP experience. Twenty-seven (90.00%) NPs had prescribing au- thority, with 21 (70.00%) having less than 2 years of prescribing experience.

The NPs’ median RN experience was 29.50 years (interquartile range [IQR]¼ 7.25, range¼ 5.00-40.00). The NPs had worked within their specialty area as an RN for a median of 15.50 years (IQR ¼ 9.50, range ¼ 5.00-39.00) before registering as an NP. Most NPs (n ¼ 27, 90.00%) completed their master’s degree in NZ.

Residents Nine female (56.25%) and 7 male (43.75%) residents participated in the study. They had an average of 6.30 years (SD ¼ 2.43) of resident experience. During the data collection period, they practiced in a variety of specialty areas; however, their specialty training areas included general practice (n ¼ 5, 31.25%), cardiology (n ¼ 3, 18.75%), respiratory (n ¼ 1, 6.25%), emergency care (n ¼ 2, 12.50%), gerontology (n ¼ 1, 6.25%), and general medicine (n ¼ 4, 25.00%). One resident was previously a medical specialist but was retraining in another specialty, and 4 (25.00%) had previously worked in other residency programs.

Most residents (n ¼ 13, 81.25%) had completed the first part of their current specialty training. The residents had completed a median of 1.50 years (IQR ¼ 1.00, range ¼ 0.00-5.00) before part 1 examinations and a mean of 1.80 years (SD ¼ 1.53) after part 1 examinations.

Diagnostic Reasoning Style As a group, NPs used an analytic-intuitive diagnostic reasoning style (M ¼ 160.8, SD ¼ 5.91), whereas residents favored an analytic style (M ¼ 157.2, SD ¼ 6.61). The 2-tailed independent t test revealed a trend toward NPs incorporating more system I processes in their diagnostic reasoning (t44 ¼ 1.91, P ¼ .06). Figure 1 shows the varying types of diagnostic reasoning styles used by participants. The individual groups’ intuitive/analytic reasoning scores

Volume 12, Issue 6, June 2016

Figure 1. Participants’ diagnostic reasoning style.

Table 1. Participants’ Intuitive/Analytic Reasoning Scores

NPs’ Intuitive/Analytic Reasoning Scores Residents’ Intuitive/Analytic Reasoning Scores

Percentile RanksScore Frequency, n (%) z Score Score Frequency, n (%) z Score

150 1 (3.3) �1.83 145 1 (6.3) �1.85 �1.09

152 1 (3.3) �1.49 149 1 (6.3) �1.24 �0.79

153 1 (3.3) �1.32 150 1 (6.3) �1.09 �0.63

154 1 (3.3) �1.16 151 1 (6.3) �0.94 �0.79

155 3 (10.0) �0.99 154 1 (6.3) �0.48 �0.33

156 2 (6.7) �0.82 155 2 (12.5) �0.33 �0.18

157 1 (3.3) �0.65 157 1 (6.3) �0.03 �0.03

158 3 (10.0) �0.48 158 2 (12.5) 0.12 0.12

159 1 (3.3) �0.31 159 1 (6.3) 0.27 0.27

160 1 (3.3) �0.14 162 2 (12.5) 0.73 0.43

162 2 (6.7) 0.20 165 1 (6.3) 1.18 0.73

163 2 (6.7) 0.37 166 1 (6.3) 1.33 0.88

164 2 (6.7) 0.54 169 1 (6.3) 1.79 1.03

165 1 (3.3) 0.70 1.18

166 1 (3.3) 0.87 1.33

167 1 (3.3) 1.04 1.49

168 5 (16.7) 1.21 1.64

173 1 (3.3) 2.06 2.39

www.npjournal.org The Journal for Nurse Practitioners - JNP 385

are outlined in Table 1. The scores of NPs reflecting an analytic or mostly analytic style (scores � 164) revealed an analytic diagnostic reasoning style was dominant in 70.0% (n ¼ 21) of NPs.

Because of the risk of a type II error when per- forming the independent t test, percentile ranks were applied to the data to show how far the NPs’ individual intuitive/analytic scores were from the residents’ mean score (Table 1). Percentile ranks showed 32.0% of NPs have a more analytic style of diagnostic reasoning when compared with the resident mean. This implies 68.0% of NPs have a less analytic style when compared with the resident mean.

The Mann-Whitney U test showed sex was not related to diagnostic reasoning style in the NP (U ¼ 40.00, z ¼ �0.04, P ¼ .97) or resident (t14 ¼ �0.27, P ¼ .79) groups. When diagnostic reasoning style was compared with participants’ previously reported diagnostic reasoning ability scores,2 the Spearman rho coefficient found that

Table 2. Differences Between Maxims Used by Participants

Maxim NP Mean R

When facing competing diagnoses favor the

simplest one

19.12

If you don’t know what to do, don’t do anything 19.88

Consider multiple separate diseases of a patient

when the result of the history and physical

examination are atypical of any one condition

26.97

Common things occur commonly 19.45

All bleeding eventually stops 25.70

All drugs work by poisoning some aspect of

normal physiology

24.20

Don’t order a test unless you know what you

will do with the results

22.70

Real disease declares itself 21.15

Treat the patient not the x-ray 24.48

Never worry alone, get a consultation 24.90

Never give two diagnoses when you can find

one that explains everything

20.73

If what you are doing is working, keep doing it.

If what you are doing is not working, stop doing it

26.33

Follow up everything 25.15

a Indicates statistical significance.

The Journal for Nurse Practitioners - JNP386

diagnostic reasoning style was not related to the accuracy of diagnostic reasoning in either the NP (rs ¼ �0.14, n ¼ 30, P ¼ .46) or resident (rs ¼ 0.03, n ¼ 16, P ¼ .90) groups.

Maxims Used to Guide Diagnostic Reasoning NPs and residents were questioned on their use of 13 maxims in their daily practice. The Mann- Whitney U test was used to identify differences between the 2 groups in how they identified with each maxim (Table 2). The maxims NPs mostly identified with included the following:

1. “Never worry alone; get a consultation.” 2. “If what you are doing is working, keep

doing it. If what you are doing is not working, stop doing it.”

3. “Follow up everything.” 4. “Consider multiple separate diseases of

a patient when the result of the history and physical examination are atypical of any 1 condition.”

ank Resident Mean Rank Significance

31.72 U ¼ 108.5, z ¼ �3.16, P ¼ .002a

30.28 U ¼ 131.5, z ¼ �2.77, P ¼ .006a

17.00 U ¼ 136.0, z ¼ �2.55, P ¼ .01a

31.09 U ¼ 118.5, z ¼ �3.17, P ¼ .002a

19.38 U ¼ 174.0, z ¼ �1.58, P ¼ .12

22.19 U ¼ 219.0, z ¼ �0.50, P ¼ .62

25.00 U ¼ 216.0, z ¼ �0.64, P ¼ .53

27.91 U ¼ 169.5, z ¼ �1.74, P ¼ .08

21.66 U ¼ 210.5, z ¼ �0.76, P ¼ .45

20.88 U ¼ 198.0, z ¼ �1.26, P ¼ .20

28.69 U ¼ 157.0, z ¼ �1.98, P ¼ .05

18.19 U ¼ 155.0, z ¼ �2.17, P ¼ .03a

20.41 U ¼ 190.5, z ¼ �1.27, P ¼ .20

Volume 12, Issue 6, June 2016

5. “Treat the patient not the x-ray.” 6. “Do not order a test unless you know what

you will do with the results.” 7. “Common things occur commonly.” The Fisher exact test analyzed how residents

differed from NPs in the way they identified with these 7 maxims (Table 3). The residents only identified with 6 of these maxims, with only 50.0% (n ¼ 8) of residents identifying with the maxim “consider multiple separate diseases of a patient when the result of the history and physical examination are atypical of any one condition.” Residents were more likely to identify with the maxim “real disease declares itself,” with 56.3% (n ¼ 9) of residents identifying with this maxim, compared with 30.0% (n ¼ 9) of NPs (Fisher exact test, P ¼ .12). The 2-tailed independent t test showed no difference in how frequently NPs and residents identified with these maxims in their daily practice (t44 ¼ �0.89, P ¼ .38).

Spearman rho indicated no relationship between participants’ identification with maxims and diag- nostic reasoning style in either the NP (rs ¼ 0.10, n ¼ 30, P ¼ .61) or resident (rs ¼ 0.38, n ¼ 16, P ¼ .15) groups. Identifying with these maxims was not related to participants’ previously reported diagnostic reasoning ability scores2 in either the NP (rs ¼ �0.17, n ¼ 30, P ¼ .37) or resident (rs ¼ �0.08, n ¼ 16, P ¼ .77) groups.

Table 3. Maxims Nurse Practitioners Identified With Most

Maxim

Never worry alone, get a consultation

If what you are doing is working, keep doing it.

If what you are doing is not working, stop doing it

Follow up everything

Consider multiple separate diseases of a patient when the resu

the history and physical examination are atypical of any one c

Treat the patient not the x-ray

Don’t order a test unless you know what you will do with the r

Common things occur commonly

a Indicates statistical significance.

www.npjournal.org

DISCUSSION The analysis identified NPs, when compared with residents, incorporated more system I processes in their diagnostic reasoning; however, both NPs’ and residents’ identification with frequently used maxims was similar. Participants’ diagnostic reasoning style and identification with maxims were not related to their diagnostic reasoning ability scores reported in Pirret et al2; although both groups identified with certain maxims and therefore used heuristics, this indicates they successfully triggered system II processes to analyze the complex case. However, Pirret et al2 identified more diagnostic reasoning errors were made by participants completing the case in the fastest times, suggesting some participants likely used system I processes when system II processes were required.

The system II processes favored by residents when measuring their diagnostic reasoning style may reflect their training system. System II processes are developed through formal training15; thus, residents participating in formal specialist training programs are likely to reflect the diagnostic reasoning style of their training system. Although all NPs completed a master’s degree, their numerous years of RN experience may mean they are less exposed to, and influenced by, solely system II approaches. It is now recognized that diagnostic reasoning requires both system I and II processes.15,16

NP

Frequency

Resident

Frequency

Significancen (%) n (%)

28 (93.3) 14 (87.5) FET, P ¼ .60

27 (90.0) 13 (81.3) FET, P ¼ .41

27 (90.0) 12 (75.0) FET, P ¼ .22

lt of

ondition

25 (83.4) 8 (50.0) FET, P ¼ .04a

25 (83.3) 14 (87.6) FET, P ¼ 1.0

esults 24 (80.0) 16 (100.0) FET, P ¼ .08

21 (70.0) 16 (100.0) FET, P ¼ .02a

The Journal for Nurse Practitioners - JNP 387

System I processes are necessary to ensure clini- cians manage their clinical workload,16 and, in most cases, it provides the correct diagnoses.7

When used by experienced clinicians, system I processes reduce the need for clinicians to ask unnecessary questions and order unnecessary diagnostic tests.9 It is when system I processes are used inappropriately that diagnostic error occurs.7

Clinicians are often unaware of the effect heuris- tics, biases, and contextual factors, such as patients’ social circumstances and clinicians’ overconfidence, emotions, and fatigue, have on diagnostic accuracy.11

System I processes are strongly affected by contextual factors,11 meaning the risk of diagnostic error increases when heuristics, biases, and contextual factors are combined.

More recently, researchers are proposing diag- nostic errors may be less about heuristics and biases and more about knowledge, experience, and clinical expertise.3,8 Experts with a high level of specialty knowledge and clinical expertise still make diagnostic errors; however, they are better at recovering from them. Error recovery requires expert knowledge, and if that knowledge is not applied, these errors can negatively impact on patients’ progress or outcome.21

Croskerry15 argues clinicians need to understand the complexity of their diagnostic reasoning and how heuristics and biases affect their individual diagnostic reasoning accuracy. With an international focus on reducing diagnostic error, it is now timely for educators preparing NPs and residents to consider including system I and II processes in training curriculums. This will make new clinicians aware of factors that contribute to diagnostic error and strategies that reduce it.

LIMITATIONS The analysis reported in this article has a number of limitations. The instrument and questionnaire relied on self-reporting and may reflect perceived rather than actual diagnostic reasoning behaviors. Difficulty in recruiting residents meant the resident group was small, which makes it difficult to generalize the results to a wider resident population. The study was inadequately powered for the 2-tailed between-group independent t test. Because this test identified a trend toward

The Journal for Nurse Practitioners - JNP388

NPs using more system I processes than residents, percentage ranks were applied to the data to aid in the interpretation of the results. The academic and registration requirements for NPs in NZ mean the results of this analysis may not reflect the diagnostic reasoning styles and identification with maxims by NPs in countries with differing academic and regis- tration requirements.

Because little research is available on maxims used by NPs and residents, the study did not use a previously validated questionnaire; rather, it focused on exploring if NPs and residents identified with them when performing their diagnostic reasoning roles.

CONCLUSION The analysis showed NPs, when compared with residents, trended toward including more system I processes in their diagnostic reasoning; however, both groups identified with maxims similarly. Diagnostic reasoning style and identification with maxims did not influence their diagnostic accuracy of a complex case, suggesting both NPs and residents triggered system II processes when required. With an international focus on reducing diagnostic error, the results of this study, highlighting that NPs and residents use system I and II processes, provide an opportunity for NPs and those involved in NP training to reflect on NPs’ diagnostic reasoning styles and how it contributes to diagnostic error and error recovery.

References

1. Gagan MJ, Boyd M, Wysocki K, Williams DJ. The first decade of nurse

practitioners in New Zealand: a survey of evolving practice. J Am Assoc

Nurse Pract. 2014;26(11):612-619.

2. Pirret AM, Neville SJ, La Grow SJ. Nurse practitioner versus doctors

diagnostic reasoning in a complex case presentation to an acute tertiary

hospital: a comparative study. Int J Nurs Stud. 2015;52(3):716-726.

3. McLaughlin K, Eva KW, Norman GR. Reexamining our bias against heuristics.

Adv Health Sci Educ. 2014;19(3):457-464.

4. Singh H, Meyer AN, Thomas EJ. The frequency of diagnostic errors in

outpatient care: estimations from three large observational studies involving

US adult populations. BMJ Qual Saf. 2014;23(9):727-731.

5. Pelaccia T, Tardif J, Triby E, Charlin B. An analysis of clinical reasoning

through a recent and comprehensive approach: the dual-process theory.

Med Educ Online. 2011;16(5890).

6. Stolper E, Van De Wiel M, Van Royen P, Van Bokhoven M, Van Der Weijden T,

Dinant GJ. Gut feelings as a third track in general practitioners’ diagnostic

reasoning. J Gen Intern Med. 2011;26(2):197-203.

7. Elia F, Aprà F, Verhovez A, Crupi V. “First, know thyself”: cognition and

error in medicine. Acta Diabetol. 2015 May 5 [Epub ahead of print].

8. Sherbino J, Norman GR. Reframing diagnostic error: maybe it’s content, and

not process, that leads to error. Acad Emerg Med. 2014;21(8):931-933.

9. Norman GR, Eva KW. Diagnostic error and clinical reasoning. Med Educ.

2010;44(1):94-100.

Volume 12, Issue 6, June 2016

10. Levine D, Bleakley A. Maximising medicine through aphorism. Med Educ.

2012;46(2):153-162.

11. Lucchiari C, Pravettoni G. Cognitive balanced model: a conceptual scheme

of diagnostic decision making. J Eval Clin Pract. 2012;18(1):82-88.

12. Durham CO, Fowler T, Kennedy S. Teaching dual-process diagnostic

reasoning to doctor of nursing practice students: problem-based learning

and the illness script. J Nurs Educ. 2014;53(11):646-650.

13. Pirret AM. Nurse practitioner diagnostic reasoning. Massey University. http://

mro.massey.ac.nz/handle/10179/4929. 2013. Accessed August 31, 2014.

14. Young ME, Brooks LR, Norman GR. The influence of familiar non-diagnostic

information on the diagnostic decisions of novices. Med Educ. 2011;45(4):

407-414.

15. Croskerry P. A universal model of diagnostic reasoning. Acad Med. 2009;

84(8):1022-1028.

16. Elstein AS. Thinking about diagnostic thinking: a 30-year perspective. Adv

Health Sci Educ Theory Pract. 2009;14(suppl 1):7-18.

17. Lauri S, Salantera S. Developing an instrument tomeasure and describe clinical

decision making in different nursing fields. J Prof Nurs. 2002;18(2):93-100.

18. Bernstein M. Medical Maxims, Pearls and Principles: Do You Think Your

doctor Is Using Any of These? If You Were a Doctor Would You? Los Angeles,

CA; 2009. http://bioethicsdiscussion.blogspot.co.nz/2009/08/medical-maxims

-pearls-and-principles-do.html. Accessed July 31, 2014.

19. Gorman D. The nurse practitioner provides a substantive opportunity for task

substitution in primary care. J Prim Health Care. 2009;1(2):142-143.

www.npjournal.org

20. IBM Corp. IBN SPSS statistics for Windows. Version 19.0. Armonk, NY: IBM

Corp; 2010.

21. Patel VL, Cohen T, Murarka T, et al. Recovery at the edge of error: debunking

the myth of the infallible expert. J Biomed Inform. 2011;44(3):413-424.

Alison M. Pirret, PhD, NP, is a senior lecturer at the school of nursing at the College of Health, Massey University and a nurse practitioner in the critical care complex at Middlemore Hospital in Auckland, New Zealand. She can be reached at A.M.Pirret@ massey.ac.nz. In compliance with national ethical guidelines, the author reports no relationships with business or industry that would pose a conflict of interest.

1555-4155/16/$ see front matter

© 2016 Elsevier Inc. All rights reserved.

http://dx.doi.org/10.1016/j.nurpra.2016.02.006

The Journal for Nurse Practitioners - JNP 389

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.