Testing Practices, Interpretation, and Diagnostic

Original Investigation | Health Policy

Testing Practices, Interpretation, and Diagnostic Evaluationof Iron Deficiency Anemia by US Primary Care PhysiciansAndrew J. Read, MD, MS; Akbar K. Waljee, MD, MS; Jeremy B. Sussman, MD, MS; Hardeep Singh, MD, MPH; Grace Y. Chen, MD, PhD;Sandeep Vijan, MD, MS; Sameer D. Saini, MD, MS

Abstract

IMPORTANCE Recognition of iron deficiency anemia (IDA) is important to initiate timely evaluationfor gastrointestinal tract cancer. Retrospective studies have reported delays in diagnostic evaluationof IDA as a common factor associated with delayed diagnosis of colorectal cancer.

OBJECTIVE To assess how US primary care physicians (PCPs) approach testing for anemia, interpretiron laboratory studies, and refer patients with IDA for gastrointestinal endoscopy.

DESIGN, SETTING, AND PARTICIPANTS This survey study, conducted in August 2019, includedmembers of the American College of Physicians Internal Medicine Insiders Panel, a nationallyrepresentative group of American College of Physicians membership, who self-identified as PCPs.Participants completed a vignette-based survey to assess practices related to screening for anemia,interpretation of laboratory-based iron studies, and appropriate diagnostic evaluation of IDA.

MAIN OUTCOMES AND MEASURES Descriptive statistics based on survey responses wereevaluated for frequency of anemia screening, correct interpretation of iron laboratory studies, andproportion of patients with new-onset IDA referred for gastrointestinal tract evaluation.

RESULTS Of 631 PCPs who received an invitation to participate in the survey, 356 (56.4%)responded and 31 (4.9%) were excluded, for an adjusted eligible sample size of 600, yielding 325completed surveys (response rate, 54.2%). Of the 325 participants who completed surveys, 180(55.4%) were men; age of participants was not assessed. The mean (SD) duration of clinicalexperience was 19.8 (11.2) years (range, 1.0-45.0 years). A total of 250 participants (76.9%) screenedat least some patients for anemia. Interpretation of iron studies was least accurate in a scenario of aborderline low ferritin level (40 ng/mL) with low transferrin saturation (2%); 86 participants (26.5%)incorrectly responded that this scenario did not indicate IDA, and 239 (73.5%) correctly identifiedthis scenario as IDA. Of 312 participants, 170 (54.5%) recommended bidirectional endoscopy (upperendoscopy and colonoscopy) for new IDA for women aged 65 years; of 305 respondents, 168 (55.1%)recommended bidirectional endoscopy for men aged 65 years.

CONCLUSIONS AND RELEVANCE In this survey study, US PCPs’ self-reported testing practices foranemia suggest overuse of screening laboratory tests, misinterpretation of iron studies, andunderuse of bidirectional endoscopy for evaluation of new-onset IDA. Both misinterpretation of ironstudies and underuse of bidirectional endoscopy can lead to delayed diagnosis of gastrointestinaltract cancers and warrant additional interventions.

JAMA Network Open. 2021;4(10):e2127827. doi:10.1001/jamanetworkopen.2021.27827

Key PointsQuestion How do US primary care

physicians approach testing for anemia,

interpretation of iron studies, and

diagnostic evaluation for iron

deficiency anemia?

Findings In this survey study of 325 US

primary care physicians, 76.9% of

respondents self-reported screening

patients for anemia, 73.5% correctly

diagnosed iron deficiency anemia, and

54.5% recommended bidirectional

endoscopy for new-onset iron

deficiency anemia among women aged

65 years (55.1% for men aged 65 years).

Meaning Screening for anemia was

commonly self-reported by primary care

physicians despite the absence of

guidelines recommending routine

anemia screening, and bidirectional

endoscopy was underused to evaluate

new-onset iron deficiency anemia.

+ Supplemental content

Author affiliations and article information arelisted at the end of this article.

Open Access. This is an open access article distributed under the terms of the CC-BY License.

JAMA Network Open. 2021;4(10):e2127827. doi:10.1001/jamanetworkopen.2021.27827 (Reprinted) October 1, 2021 1/12

Downloaded From: https://jamanetwork.com/ on 05/11/2022

https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamanetworkopen.2021.27827&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2021.27827


Introduction

Iron deficiency anemia (IDA) is a classic early diagnostic sign of gastrointestinal (GI) tract malignantneoplasm.1,2 Between 1% and 10% of adult patients with IDA may have undiagnosed GI tractcancer.3,4 Prompt and thorough evaluation of IDA is essential. However, retrospective studies havereported that delays in diagnostic evaluation of IDA are common5,6 and can lead to a delayeddiagnosis of colorectal cancer.7,8 This problem is of increasing importance given the increasingincidence of colorectal cancer among younger patients outside the age range for which routinecolorectal cancer screening is recommended, for whom IDA may be the first diagnostic sign ofmalignant neoplasm.9,10

Clinical practices related to identification and evaluation of IDA likely vary substantially, with atleast 30 different guidelines from 10 specialty societies.11-14 However, variation related to testingpractices, interpretation of laboratory test results, and diagnostic evaluation of IDA among physiciansis not well studied even though these factors may be associated with diagnostic delays, missedcancer diagnoses, and low-value care. For instance, because of a low diagnostic yield, anemiascreening is not recommended for adults except during pregnancy, and recommendations aboutscreening during pregnancy are mixed.15-17 Inappropriate screening may be associated with healthcare waste and potential unintended harm (eg, additional downstream diagnostic tests). Inaccuratetest interpretation of iron studies (eg, serum ferritin level and iron saturation) can also lead to medicalerror, with misclassification of patients’ iron level. Interpretation can be complicated because theferritin level may be elevated in the context of an inflammatory process, potentially maskingunderlying iron deficiency.2,18 Guidelines consistently recommend that men with IDA andpostmenopausal women, in the absence of obvious causes, should be evaluated for potential occultGI tract blood loss with colonoscopy and esophagogastroduodenoscopy (EGD).11,12 However, olderguidelines, which predated the increasing incidence of colorectal cancer among younger patientsobserved over the past decade, recommended against initial endoscopic evaluation forpremenopausal women with IDA given the more common cause of iron deficiency secondary tomenstrual blood loss.11 Together, these issues introduce ambiguity and complexity to the evaluationof IDA.

Our objective was to understand how primary care physicians (PCPs) approach testing andevaluation of patients for IDA. To do this, we conducted a survey study consisting of a vignette-basedassessment of PCPs to examine their practices related to IDA testing and evaluation for male andfemale patients of different ages. A better understanding of current PCP evaluation and referralpatterns for GI endoscopy in the context of IDA may reveal potential opportunities to intervene andpromote earlier diagnosis of GI tract cancers.

Methods

Study Population and Survey DistributionFor this survey study, we conducted a national online survey of internal medicine PCPs in August2019 using the American College of Physicians (ACP) Internal Medicine Insiders Panel. The ACPInsiders Panel is a nationally representative group of internal medicine physicians selected throughstratified random sampling as representative of the ACP’s membership as a whole.19 The ACP is thelargest internal medicine specialty organization, with 163 000 members. Potentially eligibleparticipants (identified as PCPs who had completed training within the ACP Insiders Paneldemographics profile) received an email invitation to complete the electronic survey, and those whocompleted the survey received points that could be redeemed for Amazon.com gift cards. Two emailreminders were sent during a 2-week eligibility period. All respondents had to meet the followinginclusion criteria: (1) practice outpatient primary care medicine, (2) practice primarily general internalmedicine and not a subspecialty, and (3) have completed residency training. The study was deemedexempt from the requirement for informed consent by the University of Michigan institutional review

JAMA Network Open | Health Policy Diagnostic Testing and Evaluation of Iron Deficiency Anemia by US PCPs



board owing to minimal potential harm related to completion of this survey study because nopatient- or physician-identifiable information was collected. The study followed the AmericanAssociation for Public Opinion Research (AAPOR) reporting guideline.

Questionnaire DesignWe developed a survey to assess current PCP practices for the testing and evaluation of IDA. Clinicalvignettes provided with the survey questions were used to address 3 sequential domains to elicitphysicians’ clinical practices and decision-making processes20: (1) the practice of obtaining acomplete blood count (CBC) for asymptomatic healthy individuals (ie, screening for anemia), (2) PCPordering and interpretation of iron laboratory studies, and (3) evaluation and management of IDAand referral for GI endoscopy (colonoscopy and/or EGD). For screening, 5 scenarios were used tounderstand testing practices for 5 hypothetical asymptomatic individuals presenting as new patients:a 35-year-old man, a 35-year-old woman (not pregnant), a 35-year-old pregnant woman, a65-year-old man, and a 65-year-old woman. For interpretation of iron laboratory studies, a series of4 combinations of ferritin level and transferrin saturation were presented: low transferrin saturation(6%) and a low ferritin level (11 ng/mL) (to convert to μg/L, multiply by 1.0), normal transferrinsaturation (30%) and normal ferritin level (150 ng/mL), normal transferrin saturation (25%) and lowferritin level (6 ng/mL), and low transferrin saturation (2%) and normal to low ferritin level (40ng/mL). Respondents selected whether the scenario represented anemia due to iron deficiency oranother cause.

Four hypothetical patient scenarios were presented for evaluation of management of newlydiagnosed IDA to assess use of EGD and colonoscopy: a 35-year-old man, a 35-year-old woman (notpregnant), a 65-year-old man, and a 65-year-old woman. We also developed and included 2questions regarding general attitudes about diagnostic testing: 1 focused on watching and waitingand 1 focused on laboratory tests and diagnostic procedures.21 The questionnaire included a subsetof questions related to knowledge and application of the 2018 American Cancer Society ColorectalCancer Screening Guideline,22 and these questions were analyzed separately.23 The questionnairewas developed in consultation with a survey development team and iteratively modified based onfeedback from 10 PCPs at the University of Michigan, Ann Arbor. The final version of thequestionnaire is available in the eAppendix in the Supplement.

Statistical AnalysisSurvey responses are summarized using proportions, and summary statistics are reported forcontinuous variables. A 2-sided P < .05 was considered statistically significant. We used a χ2 test foranalysis of the categorical variables. Statistical analysis was performed using SAS, version 9.4 (SASInstitute). Graphs were produced using GraphPad Prism, version 8.0 (GraphPad Software).

Results

Respondent CharacteristicsThe survey was distributed electronically via email invitation to 633 individuals on the ACP InsidersPanel in August 2019 who met prescreening criteria as PCPs (according to the ACP Insiders Paneldemographics profile); 2 could not be reached. Of 631 PCPs who received an invitation, 356 (56.4%)responded and 31 (4.9%) were excluded, for an adjusted eligible sample size of 600, yielding 325completed surveys (response rate, 54.2%). Of the 325 respondents who completed the survey, 180(55.4%) listed their gender as male and 315 (96.9%) were board certified in internal medicine; age ofrespondents was not assessed (Table). Respondents represented a wide variation of clinical practiceexperience. Clinical practice duration ranged from 1.0 to 45.0 years (mean [SD], 19.8 [11.2] years);231 respondents (71.1%) practiced exclusively outpatient primary care, and 120 (36.9%) had anacademic affiliation to a medical school. Of 325 physicians, 303 (93.2%) agreed that too manylaboratory tests and diagnostic procedures are performed in the US health care system.




https://www.aapor.org/Publications-Media/AAPOR-Journals/Standard-Definitions.aspx


Demographic information from the ACP Insiders Panel was analyzed to compare potentialdifferences between respondents and nonrespondents. Demographic data from August 2019 wereavailable for 454 of the 633 participants (71.7%) who were invited to complete the survey, including240 who responded to the survey and 214 who did not respond. There were no statisticallysignificant differences between respondents and nonrespondents by geographic region (West,

Table. Characteristics of 325 Respondents to the Survey

Characteristic Respondentsa

Time in clinical practice, y

Mean (range) 19.8 (1.0-45.0)

Median (IQR) 20.0 (10.0-29.0)

Region

Urban 129 (39.7)

Suburban 167 (51.4)

Rural 29 (8.9)

Gender

Male 180 (55.4)

Female 131 (40.3)

Prefer not to answer 14 (4.3)

Race and ethnicity

Asian 80 (24.6)

Black or African American 6 (1.9)

White 193 (59.4)

Otherb 7 (2.2)

Prefer not to answer 39 (12.0)

Affiliated with a medical school

Yes 120 (36.9)

No 205 (63.1)

Board-certified in internal medicine

Yes 315 (96.9)

No 10 (3.1)

Practice setting

Single-specialty office 125 (38.5)

Multispecialty office 90 (27.7)

Medical school or academic medical center 34 (10.5)

US governmentc 23 (7.1)

Hospital based 21 (6.5)

Free-standing ambulatory care or urgent care center 10 (3.1)

Institutiond 5 (1.5)

Other 17 (5.2)

Type of practice where most time is spent

All outpatient 231 (71.1)

Primarily outpatient with some inpatient 75 (23.1)

Primarily inpatient with some outpatient 10 (3.1)

Equal outpatient and inpatient 9 (2.8)

General attitudes toward testing

Reluctant to watch and wait 221 (68.0)

Overuse of laboratory tests and diagnostic procedures 303 (93.2)

Abbreviation: IQR, interquartile range.a Data are presented as number (percentage) of respondents unless otherwise

indicated.b Self-identified as other than the listed choices.c Includes Veterans Affairs and military settings.d Includes prisons, nursing facilities, and other institutions.




Midwest, Northeast, or South), age group (<40 years, 40-55 years, or >55 years), specialty (generalinternal medicine, subspecialist, hospitalist, or other), or career stage (graduated medical school >16years ago or �16 years ago). Respondents were more likely than nonrespondents to be White thana member of another racial or ethnic group (race and ethnicity were analyzed as dichotomousvariables as collected by the ACP Insiders Panel team based on participant self-report) (146 of 215White respondents [67.9%] vs 104 of 195 White nonrespondents [53.3%]; P = .003) and were morelikely than nonrespondents to be male (141 of 236 male respondents with gender data [59.7%] vs102 of 207 male nonrespondents with gender data [49.3%]; P = .03).

Screening for AnemiaRespondents were asked about their practice of testing asymptomatic healthy patients for anemia byobtaining CBCs (screening) in a series of hypothetical scenarios involving 35-year-old or 65-year-oldmen and women presenting as new patients to their practices. Obtaining screening CBCs was acommon practice among respondents; 250 of 325 physicians (76.9%) reported testing at least someasymptomatic patients aged 35 years or 65 years (Figure 1). Of 325 PCPs, 159 (48.9%) reported theywould test all these new patients for anemia by obtaining a CBC. Physicians who reported beinggenerally reluctant to recommend watching and waiting were significantly more likely to recommendscreening all the groups for anemia (odds ratio [OR], 4.23; 95% CI, 2.53-7.07; P < .001). Physicians’gender, practice region (urban, suburban, or rural), practice type (exclusively outpatient practice vssome inpatient practice), and stage in clinical practice (<5 years or �5 years) were not significantlyassociated with reporting screening. Physicians for whom the primary practice setting was anacademic medical center were less likely to recommend obtaining screening CBCs (OR, 0.40; 95% CI,0.18-0.86; P = .01). In addition, when PCPs were asked about whether they recommended 1screening vs repeated or periodic screening, more than 40% recommended repeated or periodicscreening for anemia in 65-year-old patients (132 of 324 [40.7%] for men and 133 of 324 [41.0%] forwomen) (Figure 1).

Because several guidelines recommend screening for anemia during pregnancy, an additionalscenario of a 35-year-old pregnant woman was included to compare with the scenario of the35-year-old woman who was not pregnant. For the pregnancy scenario, 105 of 325 PCPs (32.3%)reported they did not provide care to pregnant patients within their practices, deferring care ofpregnant patients to obstetrics and gynecology specialists. Of the 220 PCPs who reported theywould see pregnant patients in their practice, 188 (85.5%) reported they would screen a pregnantpatient for anemia during her first trimester, as recommended in guidelines from the Centers for

Figure 1. Rates of Obtaining Complete Blood Counts (CBCs) for Detection of Anemia in Asymptomatic Patientsby Age, Sex, and Pregnancy Status for Healthy Patients Establishing Care With a Primary Care Physician

100

80

60

40

20

0

Resp

onde

nts,

%

Managed 35 y

Womanaged 35 y

Managed 65 y

Womanaged 65 y

Pregnant womanaged 35 y

0 CBC

1 CBC

>1 CBC

Data are based on survey responses of 325 primarycare physicians reporting recommendations of 1 CBC,repeated CBCs, or no CBC. For pregnant women, therewere 220 respondents because primary carephysicians who did not provide care to pregnantwomen were excluded.




Disease Control and Prevention and the American College of Obstetricians and Gynecologists but notin the 2015 US Preventive Services Task Force updated guidelines.15-17

Interpretation of Laboratory StudiesTo assess PCPs’ choices for additional laboratory tests to be performed for evaluation of IDA,respondents were presented with a hypothetical scenario of a 65-year-old man reporting generalizedfatigue who had a hemoglobin level of 10.4 g/dL (to convert to g/L, multiply by 10.0) and meancorpuscular volume of 72.3 fL (to convert to μm3, divide by 1.0). Measurement of ferritin level, themost specific marker of iron deficiency, was the most common laboratory test recommended byrespondents (302 of 325 [92.9%]), followed by total iron-binding capacity and transferrin saturation(278 of 325 [85.5%]).

To assess PCPs’ interpretation of iron laboratory studies, respondents were presented with aseries of laboratory values and asked to determine whether the anemia resulted from iron deficiencyor an alternate cause. For scenarios in which ferritin level and transferrin saturation wereconcordantly low or normal, respondents correctly identified the cause with more than 93%accuracy: 321 of 325 respondents (98.8%) correctly attributed anemia to iron deficiency in thepresence of a low transferrin saturation (6%) and a low ferritin level (11 ng/mL), and 303 of 325(93.2%) correctly attributed anemia to an alternate cause in the presence of a normal transferrinsaturation (30%) and a normal ferritin level (150 ng/mL) (Figure 2). However, in scenarios ofdiscordant ferritin and transferrin saturation values, diagnostic accuracy decreased. For example, ina scenario with a normal transferrin saturation (25%) but low ferritin level (6 ng/mL), 44 of 325respondents (13.5%) misidentified the cause as something other than iron deficiency even thoughthe likelihood ratio for iron deficiency with a ferritin level less than 15 ng/mL is 51.9.18 In a scenariowith a low transferrin saturation (2%) but borderline low ferritin level (40 ng/mL), 86 of 325respondents (26.5%) did not consider IDA as the cause even though the likelihood ratio for irondeficiency with a transferrin saturation less than 5% is 10.5.18 In that scenario, 239 of 325respondents (73.5%) correctly identified the scenario as IDA. There were no statistically significantdifferences in correct interpretation of these scenarios by academic affiliation status, gender, stage inclinical practice, or practice of ordering ferritin level measurement for anemia evaluation.

Diagnostic Evaluation of IDAOf 325 respondents, 316 (97.2%) reported that PCPs, not specialists, were primarily responsible formanagement of IDA within their practice settings. Respondents were asked about which tests theywould obtain in a series of scenarios with new-onset IDA and negative celiac serologic test results in

Figure 2. Interpretation of Iron Studies for Diagnosis of Iron Deficiency Anemia in 4 Scenarios With DifferentCombinations of Ferritin Levels and Transferrin Saturation (TSAT)

100

80

60

40

20

0

Resp

onde

nts,

%

Low TSAT andlow ferritin level

Normal TSAT andnormal ferritin level

Normal TSAT andlow ferritin level

Low TSAT and normalto low ferritin level

Correct diagnosis

Incorrect diagnosis

Details of the scenarios are given in the QuestionnaireDesign subsection of the Methods section.




men and women aged 35 or 65 years. For the 65-year-old patients, nearly all respondentsrecommended initial GI endoscopic evaluation with colonoscopy and/or EGD (318 [97.9%] forwomen and 314 [96.6%] for men). Most respondents recommended performing both EGD andcolonoscopy for 65-year-old patients (170 of 312 [54.5%] for women and 168 of 305 [55.1%] for men)(Figure 3). However, some recommended colonoscopy alone (without EGD) for initial evaluation(135 of 312 [43.3%] for women and 124 of 305 [40.7%] for men), which could lead to a potentialdiagnostic delay and additional sedation episode if upper GI pathologic findings were present orcolonoscopy findings were normal. There were no statistically significant differences in the decisionto perform EGD and colonoscopy for 65-year-old patients based on respondents’ academic affiliationstatus, gender, stage in clinical practice, practice type, or availability of open or direct access EGD andcolonoscopy (without prior GI consultation).

Although respondents reported managing IDA in 65-year-old men and women similarly, theyused a less invasive approach with 35-year-old patients. Of 325 PCPs, 314 (96.6%) recommendedinitial endoscopy for a 65-year-old man, but only 250 of 323 (77.4%) recommended initial endoscopicevaluation for a 35-year-old man (Figure 3). In addition, respondents reported differences in the useof GI endoscopy for 35-year-old patients based on the patient’s sex. For example, a 35-year-old manwas more likely to undergo initial endoscopic evaluation with an EGD and colonoscopy than was a35-year-old woman (OR, 5.68; 95% CI, 1.89-17.03; P = .001). However, even for a 35-year-old-man,there was a potential for a diagnostic delay, with 73 of 323 respondents (22.6%) recommendingagainst initial endoscopic assessment. There were no statistically significant differences inrecommendations for diagnostic endoscopy for a 35-year-old man based on respondents’ academicaffiliation status, gender, stage in clinical practice, or practice type. The respondents were least likelyto recommend any initial endoscopic evaluation for a 35-year-old woman (45 of 314 [14.3%]). For a35-year-old woman, using empirical iron supplementation first was recommended by 264 of 312respondents (84.6%). Respondents’ academic affiliation status, gender, stage in clinical practice, andpractice type were not significantly associated with a decision to pursue diagnostic endoscopy for a 35-year-old woman.

Discussion

In this nationally representative survey study of internal medicine PCPs, we found substantialvariation in the management of anemia. Specifically, we found self-reported evidence of bothoveruse of screening CBCs (76.9% of respondents) and underuse of bidirectional endoscopy inpatients with IDA. Certain knowledge gaps also existed in the interpretation of discordant laboratory

Figure 3. Initial Diagnostic Evaluation Recommended by Respondents for Patients With New-Onset IronDeficiency Anemia and Negative Celiac Serologic Test Results by Age and Sex

100

80

60

40

20

0

Resp

onde

nts,

%

Man aged 65 y Woman aged 65 y Man aged 35 y Woman aged 35 y

EGD and colonoscopy

Colonoscopy only

EGD only

No endoscopy

EGD indicates esophagogastroduodenoscopy.




test results, which may contribute to diagnostic errors. Specifically, 26.5% of respondentsmisdiagnosed a scenario of IDA as anemia with an alternate cause despite a low transferrin saturation(2%) and a borderline low ferritin level (40 ng/mL). Several subtle but important nuances ininterpretation of these studies may contribute to errors, including the high specificity of a low ferritinlevel (98% specific for iron deficiency with a ferritin level �12 ng/mL),24 the potential for the ferritinlevel to be normal or elevated in inflammatory states when a patient is iron deficient,2,18 and theimportance of measuring serum iron level in a fasting state to avoid potential confounders of dietaryor supplemental iron.25,26 Given the relatively high baseline prevalence of anemia of 5.6% in the US,13

incorrect interpretation of these laboratory results may lead to a substantial number of missed ordelayed diagnoses of IDA and, by extension, potential delays in diagnosis of GI tract malignantneoplasm.

Our findings are in line with those of older studies on this topic. Prior work has found limitedclinical benefit associated with the common practice of obtaining routine CBCs for hospitalizedpatients27,28 and outpatients.29-31 But in a repeated survey study of family medicine PCPs, thefrequency of routine CBCs remained persistently high between 1978 and 2004, with 67% ofrespondents reporting that they would obtain a CBC for a 55-year-old woman in the most recentsurvey.32 Although overuse is a complex phenomenon, these data suggest that routine laboratoryassessments are ingrained within medical practice. In this context, it is notable that, to ourknowledge, there are no guidelines (eg, from Choosing Wisely or other organizations) explicitlyrecommending against screening for anemia in outpatients.17,33,34 In contrast, Choosing Wiselyexplicitly recommends against routine or repeated CBCs for hospitalized patients.33,35 Although the2015 US Preventive Services Task Force guidelines determined that there was insufficient evidenceto judge the benefits or harms of screening for anemia during pregnancy,17,34 there is a potential foradverse neurodevelopmental outcomes during pregnancy when the woman has iron deficiency36;multiple international guidelines have recommended screening for anemia during pregnancy despitelimited evidence.37 In this context of absent or conflicting recommendations, most respondents inthe present survey study nonetheless reported they would obtain screening CBCs for patients whowere pregnant or not pregnant. Prior work38,39 has also reported underuse of bidirectionalendoscopy, although not to the extent found in our study. Although previous guidelinesrecommended bidirectional endoscopy for men and postmenopausal women,11 subsequentAmerican Gastroenterological Association guidelines, published after this survey was conducted,recommend bidirectional endoscopy regardless of menopausal status for those without an alternateexplanation for IDA.12 Despite these recommendations, some physicians’ selection of colonoscopyinstead of combined EGD and colonoscopy for initial evaluation of IDA may be reflective of lack ofaccess. For example, in some practice settings, insurance reimbursements may disincentivizecombined EGD and colonoscopy, and physicians may not offer the combined procedures in a singleappointment.38,39 However, a serial diagnostic testing approach may lead to increased costs andrequire that patients undergo a second episode of sedation. In addition, if there is a substantial timeinterval between the colonoscopy and the EGD, there is the theoretical potential for a delay indiagnosis (although serial testing was not directly assessed in this survey study).

Given the potential consequences of missed or underevaluated IDA and the complexities of testinterpretation, clinicians may benefit from tools within the electronic health record to help improvediagnosis and reduce medical errors. For example, instead of displaying individual laboratory valuesin isolation, a computer-assisted diagnostic tool could be developed to display an individualizedinterpretation of the most likely cause of anemia in a patient. This would allow for more preciseinterpretation of laboratory studies for individuals, such as interpreting iron studies for a patient withinflammation. In addition, automated “trigger tools,” or detection algorithms for abnormal laboratorypatterns, may help identify cases of new-onset anemia that had been potentially missed by busyclinicians.5 After abnormalities are identified by these tools, a common set of clinical actions could bepresented (eg, for ordering endoscopic testing) to help facilitate obtainment of appropriate tests andminimize additional tasks for clinicians. Reducing unnecessary CBC ordering through education or




displaying cost of the laboratory testing alone is unlikely to be sufficient,40 but a multilevelintervention incorporating education, price information, feedback, and financial incentives hasdemonstrated success in an inpatient setting.41

LimitationsThis study has limitations. Respondents may have answered questions based on what they believedwas the correct evidence-based practice rather than what was reflective of their actual clinicalpractice, potentially introducing response bias. Other limitations include the possibility thatparticipants in the ACP Insiders Panel were more engaged with the organization and thus more awareof current evidence-based practices. Both of these biases would suggest that the knowledge gapsidentified in our study are underestimates of those seen in actual practice. The results may not applyto other professionals who provide primary care, such as family medicine physicians or advancedpractice physicians. A potential for nonresponse bias exists, and those who did not respond mayhave been less knowledgeable about the topic or may have differed in other ways related to theoutcomes of interest. However, we attempted to mitigate this bias through systematic efforts toincrease responses. In addition, we analyzed available demographic data for respondents andnonrespondents and found no significant differences in age, geographic location, or career stage;however, respondents were more likely to be White and male than were nonrespondents. Surveymethods may not be as well suited for the study of actual practice behaviors as other methods.However, our survey allowed us to indirectly assess physician knowledge and behavior with uniform,ideal clinical scenarios, which is not possible when secondary data are used.

Conclusions

In this survey study, PCPs’ self-reported testing practices for anemia suggest overuse of screeninglaboratory tests, misinterpretation of iron studies, and underuse of bidirectional endoscopy inevaluation of new-onset IDA. These findings suggest that IDA, a common clinical condition and earlysign of GI malignant neoplasm, is not uniformly understood, approached, and managed by practicingphysicians in the US. Given the potential association with GI cancers and the increasing incidence ofcolorectal cancer in younger adults, thorough evaluation of IDA is important for preventing delays indiagnosis.3,4,9 Further work should inform the design and implementation of tools to improvediagnostic evaluation of IDA and reduce the risk of potential diagnostic errors.

ARTICLE INFORMATIONAccepted for Publication: August 1, 2021.

Published: October 1, 2021. doi:10.1001/jamanetworkopen.2021.27827

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Read AJet al. JAMA Network Open.

Corresponding Author: Andrew J. Read, MD, MS, Division of Gastroenterology and Hepatology, Department ofInternal Medicine, University of Michigan, Ann Arbor, 3912 E Taubman Center, 1500 E Medical Center Dr, SPC 5362,Ann Arbor, MI 48109-5362 ([email protected]).

Author Affiliations: Department of Internal Medicine, University of Michigan, Ann Arbor (Read, Waljee, Sussman,Chen, Vijan, Saini); Institute for Healthcare Policy and Innovation, University of Michigan, Ann Arbor (Read, Waljee,Sussman, Vijan, Saini); Veterans Affairs Health Services Research and Development, Center for ClinicalManagement Research, Ann Arbor, Michigan (Waljee, Sussman, Vijan, Saini); Center for Innovations in Quality,Effectiveness and Safety, Michael E. DeBakey Veterans Affairs Medical Center and Baylor College of Medicine,Houston, Texas (Singh).

Author Contributions: Dr Read had full access to all of the data in the study and takes responsibility for theintegrity of the data and the accuracy of the data analysis.

Concept and design: Read, Waljee, Sussman, Saini.





https://jamanetwork.com/pages/cc-by-license-permissions/?utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2021.27827

mailto:[email protected]

Acquisition, analysis, or interpretation of data: Read, Sussman, Singh, Chen, Vijan, Saini.

Drafting of the manuscript: Read, Waljee.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Read, Waljee, Vijan.

Obtained funding: Read, Saini.

Administrative, technical, or material support: Read, Sussman.

Supervision: Sussman, Chen, Vijan, Saini.

Conflict of Interest Disclosures: Dr Saini reported receiving research and employment support from theDepartment of Veterans Affairs. No other disclosures were reported.

Funding/Support: This work was supported by grant NIH KL2TR002241 from the National Institutes of Health (DrRead) and grant CIN13-413 from the Veterans Affairs Health Services Research and Development Service Centerfor Innovations in Quality, Effectiveness and Safety (Dr Singh).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection,management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; anddecision to submit the manuscript for publication.

Additional Contributions: The American College of Physicians Internal Medicine Insiders Panel staff helped withsurvey distribution and analysis of survey panel demographics. Brian Zikmund-Fisher, PhD, and the University ofMichigan Center for Bioethics and Social Sciences in Medicine Working Group provided feedback in surveydevelopment. Contributors were not compensated for their work.

REFERENCES1. Camaschella C. Iron-deficiency anemia. N Engl J Med. 2015;372(19):1832-1843. doi:10.1056/NEJMra1401038

2. Lopez A, Cacoub P, Macdougall IC, Peyrin-Biroulet L. Iron deficiency anaemia. Lancet. 2016;387(10021):907-916. doi:10.1016/S0140-6736(15)60865-0

3. Rockey DC, Cello JP. Evaluation of the gastrointestinal tract in patients with iron-deficiency anemia. N Engl JMed. 1993;329(23):1691-1695. doi:10.1056/NEJM199312023292303

4. Stephens MR, Hopper AN, White SR, et al. Colonoscopy first for iron-deficiency anaemia: a Numbers Needed toInvestigate approach. QJM. 2006;99(6):389-395. doi:10.1093/qjmed/hcl053

5. Murphy DR, Laxmisan A, Reis BA, et al. Electronic health record-based triggers to detect potential delays incancer diagnosis. BMJ Qual Saf. 2014;23(1):8-16. doi:10.1136/bmjqs-2013-001874

6. Murphy DR, Meyer AND, Vaghani V, et al. Development and validation of trigger algorithms to identify delays indiagnostic evaluation of gastroenterological cancer. Clin Gastroenterol Hepatol. 2018;16(1):90-98. doi:10.1016/j.cgh.2017.08.007

7. Singh H, Daci K, Petersen LA, et al. Missed opportunities to initiate endoscopic evaluation for colorectal cancerdiagnosis. Am J Gastroenterol. 2009;104(10):2543-2554. doi:10.1038/ajg.2009.324

8. Singh H, Giardina TD, Petersen LA, et al. Exploring situational awareness in diagnostic errors in primary care.BMJ Qual Saf. 2012;21(1):30-38. doi:10.1136/bmjqs-2011-000310

9. Siegel RL, Fedewa SA, Anderson WF, et al. Colorectal cancer incidence patterns in the United States, 1974-2013.J Natl Cancer Inst. 2017;109(8):djw322. doi:10.1093/jnci/djw322

10. Siegel RL, Torre LA, Soerjomataram I, et al. Global patterns and trends in colorectal cancer incidence in youngadults. Gut. 2019;68(12):2179-2185. doi:10.1136/gutjnl-2019-319511

11. Goddard AF, James MW, McIntyre AS, Scott BB; British Society of Gastroenterology. Guidelines for themanagement of iron deficiency anaemia. Gut. 2011;60(10):1309-1316. doi:10.1136/gut.2010.228874

12. Ko CW, Siddique SM, Patel A, et al. AGA clinical practice guidelines on the gastrointestinal evaluation of irondeficiency anemia. Gastroenterology. 2020;159(3):1085-1094. doi:10.1053/j.gastro.2020.06.046

13. Le CH. The prevalence of anemia and moderate-severe anemia in the US population (NHANES 2003-2012).PLoS One. 2016;11(11):e0166635. doi:10.1371/journal.pone.0166635

14. Peyrin-Biroulet L, Williet N, Cacoub P. Guidelines on the diagnosis and treatment of iron deficiency acrossindications: a systematic review. Am J Clin Nutr. 2015;102(6):1585-1594. doi:10.3945/ajcn.114.103366

15. Centers for Disease Control and Prevention. Recommendations to prevent and control iron deficiency in theUnited States. MMWR Recomm Rep. 1998;47(RR-3):1-29.

16. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 95: anemia in pregnancy.Obstet Gynecol. 2008;112(1):201-207. doi:10.1097/AOG.0b013e3181809c0d




https://dx.doi.org/10.1056/NEJMra1401038

https://dx.doi.org/10.1016/S0140-6736(15)60865-0

https://dx.doi.org/10.1056/NEJM199312023292303

https://dx.doi.org/10.1093/qjmed/hcl053

https://dx.doi.org/10.1136/bmjqs-2013-001874

https://dx.doi.org/10.1016/j.cgh.2017.08.007


https://dx.doi.org/10.1038/ajg.2009.324

https://dx.doi.org/10.1136/bmjqs-2011-000310

https://dx.doi.org/10.1093/jnci/djw322

https://dx.doi.org/10.1136/gutjnl-2019-319511

https://dx.doi.org/10.1136/gut.2010.228874

https://dx.doi.org/10.1053/j.gastro.2020.06.046

https://dx.doi.org/10.1371/journal.pone.0166635

https://dx.doi.org/10.3945/ajcn.114.103366

https://www.ncbi.nlm.nih.gov/pubmed/9563847

https://dx.doi.org/10.1097/AOG.0b013e3181809c0d

17. Siu AL; U.S. Preventive Services Task Force. Screening for iron deficiency anemia and iron supplementation inpregnant women to improve maternal health and birth outcomes: U.S. Preventive Services Task ForceRecommendation Statement. Ann Intern Med. 2015;163(7):529-536. doi:10.7326/M15-1707

18. Guyatt GH, Oxman AD, Ali M, Willan A, McIlroy W, Patterson C. Laboratory diagnosis of iron-deficiency anemia:an overview. J Gen Intern Med. 1992;7(2):145-153. doi:10.1007/BF02598003

19. Butkus R, Weissman A. Internists’ attitudes toward prevention of firearm injury. Ann Intern Med. 2014;160(12):821-827. doi:10.7326/M13-1960

20. Peabody JW, Luck J, Glassman P, Dresselhaus TR, Lee M. Comparison of vignettes, standardized patients, andchart abstraction: a prospective validation study of 3 methods for measuring quality. JAMA. 2000;283(13):1715-1722. doi:10.1001/jama.283.13.1715

21. Scherer LD, Caverly TJ, Burke J, et al. Development of the Medical Maximizer-Minimizer Scale. Health Psychol.2016;35(11):1276-1287. doi:10.1037/hea0000417

22. Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guidelineupdate from the American Cancer Society. CA Cancer J Clin. 2018;68(4):250-281. doi:10.3322/caac.21457

23. Read AJ, Waljee AK, Saini SD. A national survey of adoption of the 2018 American Cancer Society ColorectalCancer Screening Guideline in primary care. Clin Gastroenterol Hepatol. 2020;S1542-3565(20)31219-2. doi:10.1016/j.cgh.2020.08.060

24. Mast AE, Blinder MA, Gronowski AM, Chumley C, Scott MG. Clinical utility of the soluble transferrin receptorand comparison with serum ferritin in several populations. Clin Chem. 1998;44(1):45-51. doi:10.1093/clinchem/44.1.45

25. Collings R, Harvey LJ, Hooper L, et al. The absorption of iron from whole diets: a systematic review. Am J ClinNutr. 2013;98(1):65-81. doi:10.3945/ajcn.112.050609

26. Fairweather-Tait SJ, Wawer AA, Gillings R, Jennings A, Myint PK. Iron status in the elderly. Mech Ageing Dev.2014;136-137:22-28. doi:10.1016/j.mad.2013.11.005

27. Frye EB, Hubbell FA, Akin BV, Rucker L. Usefulness of routine admission complete blood cell counts on ageneral medical service. J Gen Intern Med. 1987;2(6):373-376. doi:10.1007/BF02596360

28. Mozes B, Haimi-Cohen Y, Halkin H. Yield of the admission complete blood count in medical inpatients.Postgrad Med J. 1989;65(766):525-527. doi:10.1136/pgmj.65.766.525

29. Boland BJ, Wollan PC, Silverstein MD. Yield of laboratory tests for case-finding in the ambulatory generalmedical examination. Am J Med. 1996;101(2):142-152. doi:10.1016/S0002-9343(96)80068-4

30. Rüttimann S, Clémençon D, Dubach UC. Usefulness of complete blood counts as a case-finding tool in medicaloutpatients. Ann Intern Med. 1992;116(1):44-50. doi:10.7326/0003-4819-116-1-44

31. Shapiro MF, Greenfield S. The complete blood count and leukocyte differential count: an approach to theirrational application. Ann Intern Med. 1987;106(1):65-74. doi:10.7326/0003-4819-106-1-65

32. Chacko KM, Feinberg LE. Laboratory screening at preventive health exams: trend of testing, 1978-2004. Am JPrev Med. 2007;32(1):59-62. doi:10.1016/j.amepre.2006.09.002

33. ABIM Foundation. Choosing Wisely. Accessed May 1, 2021. https://www.choosingwisely.org/

34. Cantor AG, Bougatsos C, Dana T, Blazina I, McDonagh M. Routine iron supplementation and screening for irondeficiency anemia in pregnancy: a systematic review for the U.S. Preventive Services Task Force. Ann Intern Med.2015;162(8):566-576. doi:10.7326/M14-2932

35. Morgan DJ, Pineles L, Owczarzak J, et al. Accuracy of practitioner estimates of probability of diagnosis beforeand after testing. JAMA Intern Med. 2021;181(6):747-755. doi:10.1001/jamainternmed.2021.0269

36. Georgieff MK. Iron deficiency in pregnancy. Am J Obstet Gynecol. 2020;223(4):516-524. doi:10.1016/j.ajog.2020.03.006

37. Tran K, McCormack S. Screening and Treatment of Obstetric Anemia: A Review of Clinical Effectiveness, Cost-Effectiveness, and Guidelines. Canadian Agency for Drugs and Technologies in Health; 2019.

38. El-Serag HB, Xu F, Biyani P, Cooper GS. Bundling in Medicare patients undergoing bidirectional endoscopy:how often does it happen? Clin Gastroenterol Hepatol. 2014;12(1):58-63. doi:10.1016/j.cgh.2013.07.021

39. Wang P, Hutfless SM, Shin EJ, et al. Same-day vs different-day elective upper and lower endoscopicprocedures by setting. JAMA Intern Med. 2019;179(7):953-963. doi:10.1001/jamainternmed.2018.8766

40. Sedrak MS, Myers JS, Small DS, et al. Effect of a price transparency intervention in the electronic health recordon clinician ordering of inpatient laboratory tests: the PRICE randomized clinical trial. JAMA Intern Med. 2017;177(7):939-945. doi:10.1001/jamainternmed.2017.1144




https://dx.doi.org/10.7326/M15-1707

https://dx.doi.org/10.1007/BF02598003


https://jama.jamanetwork.com/article.aspx?doi=10.1001/jama.283.13.1715&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2021.27827

https://dx.doi.org/10.1037/hea0000417

https://dx.doi.org/10.3322/caac.21457



https://dx.doi.org/10.1093/clinchem/44.1.45

https://dx.doi.org/10.1093/clinchem/44.1.45

https://dx.doi.org/10.3945/ajcn.112.050609

https://dx.doi.org/10.1016/j.mad.2013.11.005

https://dx.doi.org/10.1007/BF02596360

https://dx.doi.org/10.1136/pgmj.65.766.525

https://dx.doi.org/10.1016/S0002-9343(96)80068-4

https://dx.doi.org/10.7326/0003-4819-116-1-44

https://dx.doi.org/10.7326/0003-4819-106-1-65

https://dx.doi.org/10.1016/j.amepre.2006.09.002

https://www.choosingwisely.org/


https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamainternmed.2021.0269&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2021.27827

https://dx.doi.org/10.1016/j.ajog.2020.03.006

https://dx.doi.org/10.1016/j.ajog.2020.03.006




41. Yarbrough PM, Kukhareva PV, Horton D, Edholm K, Kawamoto K. Multifaceted intervention includingeducation, rounding checklist implementation, cost feedback, and financial incentives reduces inpatientlaboratory costs. J Hosp Med. 2016;11(5):348-354. doi:10.1002/jhm.2552

SUPPLEMENT.eAppendix. Anemia Survey




https://dx.doi.org/10.1002/jhm.2552

Documents

Testing Practices, Interpretation, and Diagnostic