Physician Characteristics and Decisions Regarding Cancer Screening: A Systematic Review

Abstract

The guidelines provided by experts regarding various cancer screening tests are not universally accepted by physicians or patients. This systematic review describes the literature regarding the associations of physician characteristics with the implementation of and referral of patients for selected cancer screening tests. In October 2016, the authors conducted a thorough search of articles found in 4 databases, using keywords describing physician characteristics and cancer screening. English-language articles reporting on patient surveys or records of patients' screening history, and surveys of physicians' practices or opinions, in the United States were included. The physician characteristics most commonly analyzed were specialty and sex. The screening tests most commonly analyzed were those for cervical and colorectal cancers. Female and white physicians were found to screen more patients than male and nonwhite physicians. Obstetrician-gynecologists screened more for breast and cervical cancer than internists, who screened more than family or general practitioners. Physician sex, race, and specialty were consistently associated with cancer screening practices and should be the focus of efforts to harmonize practices with recommendations.

Introduction

Recent studies and commentaries have cast doubt on the efficacy and utility of several cancer screening tests.^1

–4 However, many patients seem to take the benefits of screening for granted.^5

–8

Like most medical decisions, those regarding cancer screening involve the interplay and interaction of 4 groups of variables: patient demographic characteristics (eg, age,^9

–12 sex,^12

–16 race/ethnicity,^12,17,18 socioeconomic status^12,19,20), patient clinical characteristics (eg, risk of the disease in question,^21

–24 comorbidities^25
–27), system characteristics (eg, type of insurance,^20,28
–30 setting in which the patient is seen^31

–35), and physician characteristics (eg, specialty,^36
–38 age,^39
–41 sex,^39,40,42,43 board certification,^38,40,44 foreign medical graduate^{39,40,44
–46}). Studies have investigated the effects of patient and system variables on the utilization of various cancer screening tests.

This paper reviews the biomedical literature regarding the associations of physician characteristics with decisions regarding various common cancer screening tests. An understanding of these relationships may help to explain why screening practices vary^1

–4 and often are at odds with recommendations from expert panels and public health agencies.⁴⁷

Methods

Citation search strategy

The databases MEDLINE (1946 – October 2016), PubMed (1900 – October 2016), PsycINFO (1967 – October 2016), and EMBASE (1980 – October 2016) were searched using terms describing physician characteristics (keywords: physician characteristics, physician age, physician gender, physician education, physician specialty, physician training, physician experience, physician race, physician practice, international medical graduate, and board certification) and cancer screening (keywords: cancer screening, screening practices, and screening patterns). The “related articles” feature of PubMed was used to identify additional relevant articles. The study team also hand searched the references of retrieved articles to identify studies that the search strategy might have missed (Figure 1). Abstracts were included only if the data were not yet published in peer-reviewed journals.

FIG. 1.

Flow diagram of article selection based on searches of keywords related to physician characteristics and cancer screening. RCT, randomized controlled trial.

Study selection

After all studies based on the search strategy were identified, 2 authors independently reviewed titles, then abstracts, then articles to confirm potential eligibility. Studies that were non–cancer related, focused on basic research, evaluated policies or programs, did not report on physician characteristics, were written in languages other than English, or sampled populations outside the United States were excluded.

Studies that focused on physician characteristics associated with receipt of cancer screening, selected studies that used patient-level data on cancer screening patterns linked with data on attending physician characteristics, and surveys of physician attitudes, recommendations, and use of cancer screening procedures were included. For studies that examined several different preventive screening practices, only those outcomes that were cancer related were included. The few discrepancies between the 2 reviewing authors were quickly resolved by either reaching consensus or seeking an opinion from a third author.

Data extraction and synthesis

A standardized extraction form was used to obtain data on study design and relevant results. The articles selected were reviewed, and data on study design, subjects, sampling methods, sample size, screening test type, response rate, and quantitative results were recorded in table form. All data were abstracted independently by each of the 2 authors and verified for accuracy by a third.

The selected articles were categorized into 10 groups based on physician characteristics observed in the studies: age, sex, race, experience, specialty, board certification, degree, physician training location (Unite States or foreign), volume of patients, and academic affiliation. Study design was classified as prospective cohort, retrospective cohort, case control, or cross-sectional. Subjects were classified as patients or physicians. The sample description was a brief summary of the cohort source. The screening test types included breast cancer (mammogram), colorectal cancer (CRC; colorectal screening), cervical cancer (Pap smear), and prostate cancer (PSA [prostate-specific antigen]), among others. The response rate included both the percent of physicians who responded to a survey and the percent of patients ascertained. The quantitative results were the measures of association obtained from the adjusted regressions. The quantitative results of each article were classified based on the overall effects observed.

Results

The search returned 1696 articles, 42 of which were included in the analysis (see Supplementary Table S1). Twenty-two studies (52%) were surveys of physicians, 14 (33%) were medical record reviews, and 6 (14%) were patient surveys.

The physician characteristics most commonly analyzed were specialty (81%), sex (69%), race (33%), experience (31%), and age (31%). Almost half (45%) the studies analyzed screening for cervical cancer, 33% for CRC, 29% for breast cancer, and 19% for prostate cancer; 1 analyzed lung cancer screening.

Demographics

Age

The papers yielded no consensus on the association of physician age with breast cancer screening (Table 1). For mammography, 2 studies had null results,^42,48 while others found a positive relationship⁴⁰ and an inverse relationship.⁴⁹ For clinical breast examination, results showed no association⁴⁹ or that older physicians screen less often.⁴⁸ Two of the 3 studies on Pap testing for cervical cancer^48,50 found no association with physician age and 1³⁸ found an inverse relationship.

Table 1.

Summary of Findings from Articles That Analyzed the Impact of Physician Age on Cancer Screening

Author	Screening test	Measure	Results	Overall effect (OR [95% CI])
Leach et al (2012)⁴²	Mammogram	Over-recommend: would recommend mammogram alone or in combination with CBE for a woman with a terminal comorbidity	Percentages of physicians by age: <50: 44.1 ≥50: 47.6	Null (OR not reported)
Tan et al (2013)⁴⁰	Mammogram	Patients receiving mammogram	Percentages of physicians screening patients with life expectancy <7 yrs; ≥7 yrs: <50: 32.4; 56.8 ≥50 ^*: 32.6; 55.2	Younger physicians screen less (0.90 [0.86, 0.95]; 0.94 [0.90, 0.98])
Lane et al (2000)⁴⁹	Mammogram CBE	Patient self-report of (1) receiving recommendation for mammogram from a physician, (2) having mammogram in past 2 yrs, and (3) having CBE in past yr	OR [95% CI] (listed as MD recommendation; mammogram in past 2 yrs; CBE in past yr) <40^* 40–49: 0.74 [0.57, 0.97]; 0.77 [0.62, 0.96]; 0.92 [0.76, 1.12] 50–59: 0.67 [0.51, 0.89]; 0.76 [0.60, 0.97]; 0.96 [0.78, 1.18] ≥60: 0.50 [0.36, 0.69]; 0.65 [0.49, 0.87]; 0.97 [0.75, 1.26]	Older physicians screen less with mammogram than those aged <40 yrs CBE: Null
Yabroff et al (2009)³⁸	Pap test	Guideline-consistent response to clinical vignettes	OR [95% CI]: <40 ^* 40–49: 0.63 [0.42, 0.93] 50–59: 0.57 [0.37, 0.88] ≥60: 0.52 [0.30, 0.90]	Older physicians less guideline consistent
Bruder et al (2012)⁵⁰	Pap test	Physician has no intention to change Pap test frequency following HPV vaccination (guideline consistent)	Percentages: 25–39: 84.5 40–49: 82.8 ≥50: 79.7	Null (P = .356)
Weisman et al (1989)⁵⁴	CBE, Self breast exam, Mammogram, Pap test, DRE, FOBT, FS	Predictor of physician adherence to ACS recommendations	Standardized partial regression coefficient: Physician's age: −0.24	Older physicians less likely to adhere to ACS recommendations (P < .05)
Richards et al (1998)⁵¹	CRC screening	Physician recommendation for CRC screening in women	Percentages (listed as FS in women <50 yrs; COL in women of all ages): <45^* : 17; 10 45–54: 29; 23 ≥55: 47; 32	Physicians ≥55 screen more with FS (3.79 [1.99, 7.36]) Physicians ≥55 (3.42 [1.84, 5.93]) and 45–54 (2.51 [1.29, 4.95]) screen more with COL
Yabroff et al (2011)⁴⁴	CRC screening	All guideline-consistent responses to clinical vignettes	OR [95% CI]: <40^* 40–49: 0.55 [0.34, 0.88] 50–59: 0.39 [0.23, 0.67] ≥60: 0.17 [0.09, 0.35]	Older physicians are less guideline consistent (ie, screen more)
Zapka et al (2012)⁴¹	COL	Physician report of substantial changes in the past 3 yrs in the number of tests ordered or supervised	OR [95% CI]: <40 ^* 40–49: 1.1 [0.7–1.8] 50–59: 2.0 [1.3–3.1] ≥60: 2.1 [1.2–3.9]	Older physicians screen more (ie, more report increase)
Singal et al (2013)³⁹	CRC screening	Receipt of guideline-consistent CRC screening (FOBT, FS, barium enema, COL)	Percentages: ≤45: 57.3 46–65: 57.5 >65: 52.2	Older physicians screen less (white and Hispanic patients, P < .001; blacks, P = .005)
Hall et al (2011)⁵²	PSA test	Physician indication of (A) discussing prostate screening with all male patients, and (B) having a discussion policy to talk patients into getting PSA test	Percentages (listed as A, B): 31–39 ^*: 80.0; 44.7 40–49: 85.4; 64.5 50–59: 74.0; 70.3 ≥60: 80.6; 81.6	A. Null (1.06 [0.52, 2.18]; 0.67 [0.25, 1.78]; 0.74 [0.20, 2.75]) B. Older physicians promote screening more (2.47 [1.33, 4.56]; 2.88 [1.23, 6.74]; 2.03 [0.52, 7.95])
Goodwin et al (2014)⁵³	PSA test	Patients receiving PSA	Percentages (listed as 2007; 2010): ≤50^* : 31.0; 29.6 50–60: 33.8; 30.4 >60: 36.5; 32.2	Null (50–60: 1.03 [0.89, 1.19]; 0.95 [0.81, 1.11] >60: 1.15 [0.95, 1.39]; 1.03 [0.84, 1.27])
Brownson et al (1993)⁴⁸	Mammogram, CBE, Pap test, DRE, FS, FOBT, skin, oral cavity, testicular exams	Physician indication of current level of activity of screening tests (5-point Likert scale)	Correlation coefficients: Skin exam: r = −.14 CBE: r = −.16 FOBT: r = −.15	Older physicians screen less with skin exam (P = .04), CBE (P = .03), and FOBT (P = .03) Other exams: null (P not reported)

ACS, American Cancer Society; CBE, clinical breast exam; CI, confidence interval; COL, colonoscopy; CRC, colorectal cancer; DRE, digital rectal exam; FOBT, fecal occult blood test; FS, flexible sigmoidoscopy; OR, odds ratio; PSA, prostate-specific antigen; yrs, years.

Reference.

Three of 5 studies found a positive relationship between physician age and CRC screening rates.^41,44,51 These studies generally looked at a combination of CRC screening methods, including fecal occult blood test (FOBT), flexible sigmoidoscopy, and colonoscopy. Two studies found that older physicians screen less often for CRC,³⁹ but Brownson et al found that older physicians screened less often with FOBT, but not with flexible sigmoidoscopy.⁴⁸ Two studies found no association of physician age with PSA testing,^52,53 but Hall et al found that older physicians were marginally more likely to have a policy of encouraging patients to get PSA tests.

A study by Weisman et al found that older physicians were less likely than younger physicians to adhere to American Cancer Society (ACS) guidelines for breast, cervical, and CRC screening of asymptomatic patients older than 65 years, although it was unclear whether the nonadherent physicians screened more or less often than recommended.⁵⁴

Sex

Physician sex most strongly influenced screening practices for breast cancer, but also influenced gynecological and CRC screening (Table 2). Six of 10 studies found that female physicians screen more often with mammography than male physicians.^{40,42,45,49,55,56} One study found that males screen more often,⁵⁷ and 3 others showed no association.^58
–60 For clinical breast examinations (CBEs), 1 paper showed no association,⁵⁷ and another that females screen more often.⁴⁹

Table 2.

Summary of Findings from Articles That Analyzed the Impact of Physician Sex on Cancer Screening

Author	Screening test	Measure	Results	Overall effect (OR [95% CI])
Leach et al (2012)⁴²	Mammogram	Over-recommend: would recommend mammogram alone or in combination with CBE for a woman with a terminal comorbidity	Percentages: Male^* : 42.6 Female: 52.1	Females recommend screening more (1.40 [1.03, 1.90])
Tan et al (2013)⁴⁰	Mammogram	Patients receiving mammogram	Percentage of physicians screening patients with life expectancy <7 yrs; ≥7 yrs: Male ^*: 31.3; 54.9 Female: 38.2; 63.1	Females screen more (1.33 [1.25, 1.42]; 1.34 [1.28, 1.41])
Lane et al (2000)⁴⁹	Mammogram CBE	Patient self-report of (1) receiving recommendation for mammogram from a physician, (2) having mammogram in past 2 yrs, and (3) having a CBE in the past year	OR [95% CI] (listed as MD recommendation; mammogram in past 2 yrs; CBE in past yr) Female: 1.55 [1.15, 2.08]; 1.34 [1.06, 1.68]; 1.27 [1.04, 1.55]	Females screen more with mammogram and CBE
Tishler et al (2000)⁵⁷	Mammogram CBE	Patients receiving mammogram or CBE in previous 2 yrs	Percentages (listed as mammogram, CBE): Male: 89.2; 75.9 Female: 74.5; 74.5	Males screen more with mammogram (P = .045) CBE: null (P = 1.000)
Lurie et al (1993)⁵⁶	Mammogram Pap test	Patients receiving Pap test	OR [95% CI]: Female: 1.99 [1.72, 2.30] (Mammogram OR not reported)	Females screen more
Lurie et al (1997)⁵⁹	Mammogram Pap test	Physician belief in 12-month screening interval	Percentages (listed as Pap test; mammogram for women 40–50; mammogram for women over 50) Female: 83; 8; 96 Male: 72; 14; 92	Females screen more with Pap test (P ≤ .01) Mammogram: null (P > .05)
Parkerton et al (2004)⁷⁶	Mammogram Pap test	Physician performance of cancer screening (composite of mammogram and Pap test)	Standardized beta coefficient: Sex: β = −.253	Null (P ≥ .05)
Heflin et al (2006)⁵⁵	Mammogram Pap test	Over-screen: somewhat/very likely to offer mammogram to woman with life expectancy <5yr Under-screen: less likely to offer mammogram to woman with life expectancy >10 yrs Over-screen: somewhat/very likely to offer Pap test to woman with life expectancy <5 yr	Percentages (listed as over-screen with mammogram; under-screen with mammogram; over-screen with Pap test): Female: 54.7; 7.8; 27.3 Male: 40.5; 16.0; 24.0	Females more often over-screen with mammogram (P < .001) Males more often under-screen with mammogram (P = .0068) Over-screen with Pap test: null (P = .40)
Taylor et al (2004)⁶⁴	Pap test	Patient self-report of receiving Pap test in last 2 yrs	Percentages: Female ^*: 37.8 Male: 28.9	Males screen less (0.53 [0.33, 0.88])
Yabroff et al (2009)³⁸	Pap test	Guideline-consistent response to clinical vignettes	OR [95% CI]: Female: 1.31 [0.90, 1.90]	Null
Almeida et al (2013)⁶¹	Pap test	Overuse: ≥2 Pap tests for woman with a uterus, ≥1 for woman without a uterus in 3 yr period	OR [95% CI]: Female: 1.09 [0.81, 1.48]	Null
Bruder et al (2012)⁵⁰	Pap test	Physician intention not to change Pap test frequency following HPV vaccination (guideline consistent)	Percentages: Male: 79.9 Female: 84.5	Null (P = .094)
Saraiya et al (2010)⁶³	Pap test HPV DNA test	Physicians who recommend next Pap and HPV DNA tests according to guidelines	Adjusted percentages: Male ^*: 17.1 [14.5, 20.0] Female: 16.8 [13.1, 21.3]	Null (P = .89)
Stormo et al (2012)⁴³	Pelvic exam	Physician use of routine pelvic exams to screen for ovarian/gynecological cancer	Percentages (listed as use for ovarian cancer; gynecological cancer): Male ^*: 51.4; 61.9 Female: 64.1; 72.9	Females screen more (2.1 [1.6, 2.9]; 2.1 [1.5, 2.9])
Arredondo et al (2003)⁶²	Cervical cancer screening	Score indicating likelihood to address cervical cancer with hypothetical patient	Mean (SD): Male: 4.15 (0.92) Female: 4.35 (0.92)	Null (P < .30)
Pham et al (2005)⁴⁵	Mammogram CRC screening	Reception of mammogram and CRC screening for Medicare beneficiaries age ≥65 yrs	Percentages (listed as mammogram, CRC screening): ^*Male: 46.1; 9.1 Female: 50.0; 8.7	Females screen more with mammogram (1.33 [1.05, 1.68]) CRC screening: null (0.90 [0.74, 1.09])
Liang et al (2009)⁵⁸	Mammogram Pap test CRC screening (FOBT, FS, COL)	Patient self-report of physician recommendation for mammogram, Pap test, and CRC screening in past 2 yrs	Percentage (listed as: mammogram; Pap test; CRC screening): Male^* : 62; 51; 52 Female: 65; 59; 51	Null (0.91 [0.49, 1.68]; 1.22 [0.67, 2.21]; 0.78 [0.42, 1.45])
Thompson et al (2014)⁶⁰	Mammogram Pap test CRC screening	Patients receiving guideline-consistent screening for cervical, breast, and CRC	OR [95% CI] (listed as mammogram; Pap test; CRC screening): Female: 1.16 [1.00, 1.34]; 1.66 [1.51, 1.82]; 1.05 [0.91, 1.21]	Females screen more with Pap test Mammogram and CRC screening: null
Armour et al (2003)⁷³	CRC screening	Patients receiving CRC screening	OR [95% CI] (listed as CRC screening; FOBT; FS or COL): Female: 0.84 [0.64, 1.12]; 0.85 [0.61, 1.18]; 0.85 [0.58, 1.24]	Null (P = .23; P = .32; P = .37)
Shokar et al (2009)⁶⁶	CRC screening	Patient self-report of physician recommendation of CRC screening	Percentages: Male^* : 56.1 Female: 67.0	Females screen more (1.82 [1.20, 2.75])
Cooper et al (1997)⁶⁵	CRC screening	Physicians who recommend screening with FOBT or FS for patients with comorbidities	Percentages (listed as no; mild; moderate; terminal comorbidities): FOBT: Female: 93.9; 90.8; 85.4; 50.8 Male: 91.4; 84.9; 78.3; 34.3 FS: Female: 46.9; 32.3; 21.5; 6.9 Male: 50.4; 28.0; 17.8; 4.4	Females more likely to recommend screening with FOBT in patients with terminal comorbidities (P < .001) FS: null
Yabroff et al (2011)⁴⁴	CRC screening	All guideline-consistent responses to clinical vignettes	OR [95% CI]: Female: 0.94 [0.63, 1.41]	Null
Singal et al (2013)³⁹	CRC screening	Receipt of guideline-consistent CRC screening (FOBT, FS, barium enema, COL)	Percentages: Male: 56.5 Female: 59.7	Females screen more (P < .001)
Edlefsen et al (1999)⁶⁷	PSA test	Routine recommendation of PSA test	Percentages (listed as always; only if high risk; only if requested): Male: 68.7; 15.4; 9.2 Female: 62.3; 15.4; 14.6	Null (P ≥ .01)
Hall et al (2011)⁵²	PSA test	Physician indication of (A) discussing prostate screening with all male patients, and (B) having a discussion policy to talk patients into getting PSA test	Percentages (listed as A, B): Male^* : 80.1; 68.8 Female: 80.1; 52.0	Null (0.98 [0.59, 1.62]; 0.84 [0.52, 1.34])
Pollack et al (2012)²⁰	PSA test	(A) Physician indication of there being an age at which he/she typically stops screening an otherwise healthy man (B) Physician indication that he/she takes life expectancy into account when deciding not to order PSA screening	Percentages not reported because of small cell sizes and concerns over confidentiality	Null (OR not reported)
Goodwin et al (2014)⁵³	PSA test	Patients receiving PSA	Percentages (listed as 2007; 2010): Female^* : 26.3; 23.8 Male: 33.5; 30.6	Null (1.31 [0.93, 1.84]; 1.30 [0.89, 1.89])
Klabunde et al (2012)⁶⁸	Chest radiograph LDCT	Recommendation of chest radiograph or LDCT for asymptomatic patient	OR [95% CI] (listed as chest radiograph; LDCT): Female: 1.2 [0.8, 1.7]; 0.8 [0.5, 1.4]	Null
Tyler et al (2006)⁸⁰	Family risk of cancer	Detail of cancer information in notes of medical interview	Percentages (listed as inadequate, adequate, comprehensive): Female: 71.3; 21.8; 6.9 Male: 59.2; 32.1; 8.7	Null (P = .154)

AMA, American Medical Association; CBE, clinical breast exam; CI, confidence interval; COL, colonoscopy; CRC, colorectal cancer; FOBT, fecal occult blood test; FS, flexible sigmoidoscopy; LDCT, low-dose computed tomography; OR, odds ratio; PCP, primary care physician; PSA, prostate-specific antigen; yrs, years.

Reference.

Studies of gynecological cancer screening did not show an obvious pattern. Only 4 of 12 studies found that females screen more often.^43,56,59,60 Six of the 8 remaining studies found no association,^{38,50,55,58,61
–63} and 1 found that males screen more often.⁶⁴

Three of the 9 studies of CRC screening found that females screen more often^39,65,66 and 6 found no association. Four studies found no association between physician sex and prostate cancer screening discussion, recommendation, or frequency.^20,52,53,67 The one study of physician sex and lung cancer screening found no correlation.⁶⁸

Race/ethnicity

Findings varied regarding the relationship between mammography/CBE screening and physician race (Table 3). Leach et al found that non-Hispanic white physicians were less likely to over-recommend screening than physicians of other races,⁴² while another study found no association.⁴⁹ Lew et al found lower screening rates among patients of Korean physicians in the United States than among patients of non-Koreans.⁶⁹

Table 3.

Summary of Findings from Articles That Analyze the Impact of Physician Race/Ethnicity on Cancer Screening

Author	Screening test	Measure	Results	Overall effect (OR [95% CI])
Leach et al (2012)⁴²	Mammogram	Over-recommend: would recommend mammogram alone or in combination with CBE for a woman with a terminal comorbidity	Percentages: Non-Hispanic White^* : 41.2 Other: 55.5	Those of race other than non-Hispanic white screen more (1.72 [1.26, 2.35])
Lane et al (2000)⁴⁹	Mammogram CBE	Patient self-report of (1) receiving recommendation for mammogram from a physician, (2) having mammogram in past 2 yrs, and (3) having CBE in past yr	OR [95% CI] (listed as MD recommendation; mammogram in past 2 yrs; CBE in past yr): Non-Hispanic White ^* Non-Hispanic Black: 0.72 [0.49, 1.05]; 0.86 [0.61, 1.21]; 0.85 [0.62, 1.17] Hispanic: 0.63 [0.39, 1.02]; 1.35 [0.84, 2.16]; 0.95 [0.62, 1.45] Other/unknown: 0.81 [0.64, 1.04]; 0.83 [0.66, 1.03]; 0.98 [0.80, 1.19]	Null
Lew et al (2003)⁶⁹	Mammogram CBE Pap test	Patient self-report of receiving mammogram, CBE, and Pap test in past 2 yrs	Percentages (listed as mammogram; CBE; pap test): Korean: 57; 39; 56 Non-Korean: 89; 73; 72	Non-Koreans screen more than Koreans (P < .001 for each)
Taylor et al (2004)⁶⁴	Pap test	Patient self-report of receiving Pap test in last 2 yrs	Percentages: Vietnamese: 28.1 Other ^*: 39.9	Null (1.27 [0.75, 2.16])
Bruder et al (2012)⁵⁰	Pap test	Physicians who have no intention to change Pap test frequency following HPV vaccination (guideline consistent)	Percentages: Race: White: 82.1 Other: 80.7 Ethnicity: Hispanic or Latino: 84.0 Not Hispanic or Latino: 82.0	Race: null (P = .663) Ethnicity: null (P = .726)
Saraiya et al (2010)⁶³	Pap test HPV DNA test	Physicians who recommend next Pap and HPV DNA tests according to guidelines	Adjusted percentages: Non-Hispanic White ^*: 19.8 [16.8, 23.1] Non-Hispanic Asian: 13.9 [9.3, 20.7] Other: 7.8 [4.8, 12.4]	Non-Hispanic whites recommend more frequent screening than non-Hispanic Asians and those of other races (P < .001)
Stormo et al (2012)⁴³	Pelvic exam	Physicians who use routine pelvic exams to screen for ovarian/gynecological cancer	Percentages (listed as use for ovarian cancer; gynecological cancer): White^* : 59.0; 68.7 Asian: 39.0; 51.3 African-American: 57.6; 67.8 Other: 47.8; 55.2	Asians screen less for ovarian cancer than whites (0.6 [0.4, 0.9])
Arredondo et al (2003)⁶²	Cervical cancer screening	Score indicating likelihood to address cervical cancer with hypothetical patient	Mean (SD): Black: 4.67 (0.71) White: 4.28 (0.81) Asian: 3.90 (1.21)	Blacks more likely to address screening than Asians (P < .05)
Shokar et al (2009)⁶⁶	CRC screening	Patient self-report of physician recommendation of CRC screening	Percentages: Non-Hispanic White^* : 66.2 Black: 68.6 Hispanic/Latino: 37.8 Asian: 60.0	Hispanics screen less than non-Hispanic whites (0.47 [0.16, 0.94])
Zapka et al (2011)⁴⁶	CRC screening	Frequency of recommending FOBT + COL or COL alone compared to other tests or combinations	OR [95% CI]: Non-Hispanic White ^* Asian: 0.7 [0.5, 1.4] Other: 0.5 [0.3, 0.9]	Non-Hispanic whites and Asians more likely to recommend FOBT + COL or COL alone (P = .0267)
Zapka et al (2012)⁴¹	COL	Physician report of substantial changes in the past 3 yrs in the number of tests ordered or supervised	OR [95% CI]: Non-Hispanic White^* Non-Hispanic Black: 1.7 [0.7,4.2] Hispanic: 1.3 [0.5, 2.8] Asian: 1.7 [0.9, 2.9] Other: 1.6 [0.7, 3.7]	Null
Hall et al (2011)⁵²	PSA test	Physician indication of (A) discussing prostate screening with all male patients, and (B) having a discussion policy to talk patients into getting PSA test	OR [95% CI] (listed as A, B): Non-Hispanic White ^*: 77.8; 61.8 Non-Hispanic Black: 82.6; 76.5 Hispanic: 70.9; 77.5 Other: 90.4; 66.7	A. Null (1.15 [0.76, 1.74]; 0.61 [0.24, 1.57]; 2.51 [1.13, 5.58]) B. Non-Hispanic blacks screen more than non-Hispanic whites (2.80 [1.88, 4.16])
Pollack et al (2012)²⁰	PSA test	(A) Physician indication of there being an age at which he/she typically stops screening an otherwise healthy man (B) Physician indication that he/she takes life expectancy into account when deciding not to order PSA screening	White, Black, Asian, Other Percentages not reported because of small cell sizes and concerns over confidentiality	Blacks less likely to have age when they discontinued screening than non-Blacks (P = .044)
Wallner et al (2012)⁷⁰	PSA test	Patients receiving PSA screening	Percentages: White: 30.51 Asian: 30.29 Black: 28.56 Hispanic: 28.06 Other: 31.52	Whites and Asians screen more than Blacks and Hispanics (P < .001)

CBE, clinical breast exam; CI, confidence interval; COL, colonoscopy; CRC, colorectal cancer; FOBT, fecal occult blood test; OR, odds ratio; PSA, prostate-specific antigen; yrs, years.

Reference.

Several studies showed that Asian physicians screen less often for gynecological or ovarian cancers. One study found that Asians screen less often for cervical cancer than blacks,⁶² and another found that Asians screen less often for ovarian cancer than whites.⁴³ A third found that white physicians recommended Pap and HPV DNA screening more than did Asians, but that both did so more than others⁶³; another study found higher Pap screening rates among patients of non-Korean physicians than among patients of Koreans.⁶⁹ Two other studies found no association.^50,64

Two studies suggested that white physicians did more CRC screening than physicians of other races.^46,66 Zapka et al⁴⁶ found that non-Hispanic white and Asian primary care physicians were more likely than other primary care physicians to report recommending FOBT with colonoscopy/colonoscopy only, compared with other tests or test combinations. However, their 2012 study⁴¹ did not find an association of race/ethnicity with increased (over time) ordering of colonoscopy vs. other tests.

Two studies indicated that black physicians screen more than those of other races for prostate cancer. One study found that black physicians were less likely than others to discontinue annual screening of patients older than a specific age cutoff,²⁰ and another found that non-Hispanic black physicians were more likely to discuss PSA tests than non-Hispanic whites.⁵² Conversely, Wallner et al found that white and Asian providers screened more often than black and Hispanic providers.⁷⁰

Training and background characteristics

Degree

The one study to compare screening practices of allopathic and osteopathic doctors found that allopathic physicians screened more with mammograms, CBE, Pap tests, digital rectal exams, sigmoidoscopy, FOBT, skin exams, oral cavity exams, and testicular exams (data not shown).⁴⁸ Another study found no differences between allopathic physicians and physician assistants and nurse practitioners with respect to mammography, Pap test, and CRC screening.⁶⁰

Training, foreign vs. United States

Findings varied regarding the relationship between location of medical training and mammographic screening practices (Table 4). One study found no association,⁴² another found that foreign-trained physicians screen more often than US-trained physicians,⁴⁰ and a third study found that physicians trained in the United States or Canada screened more often than those trained elsewhere.⁴⁵ Findings regarding training location and CRC screening also varied. One study found no association,⁴⁵ but another found that US-trained physicians screen more often than those trained elsewhere.³⁹ Yabroff et al found that US-trained physicians were more guideline consistent than others,⁴⁴ suggesting lower screening rates, while Zapka et al⁴⁶ found that US-trained physicians were less likely than foreign-trained physicians to discuss more than 1 test option with their patients.

Table 4.

Summary of Findings from Articles That Analyzed the Impact of Physician Training Location (United States or Abroad) on Cancer Screening

Author	Screening test	Measure	Results	Overall effect (OR [95% CI])
Leach et al (2012)⁴²	Mammogram	Over-recommend: would recommend mammogram alone or in combination with CBE for a woman with a terminal comorbidity	Percentages: Foreign-trained: 49.9 US-trained: 44.4	Null (OR not reported)
Tan et al (2013)⁴⁰	Mammogram	Patients receiving mammogram	Percentage of physicians screening patients with life expectancy <7 yrs; ≥7 yrs: US-trained ^*: 28.8; 50.7 Foreign-trained: 34.0; 58.4	Foreign-trained screen more (1.30 [1.23, 1.38]; 1.35 [1.29, 1.41])
Pham et al (2005)⁴⁵	Mammogram CRC screening	Receipt of mammogram and CRC screening by Medicare beneficiaries age ≥65 yrs from US physician respondents to the 2000–2001 Community Tracking Study Physician Survey	Percentages (listed as mammogram, CRC screening): ^*US or Canada-trained: 48.2, 9.3 Foreign-trained: 39.7, 7.7	Foreign-trained screen less with mammogram (0.78 [0.63, 0.98]) CRC: null (0.92 [0.76, 1.12])
Yabroff et al (2011)⁴⁴	CRC screening	All guideline-consistent responses to clinical vignettes	OR [95% CI]: Foreign-trained^* US-trained: 1.99 [1.11, 3.58]	US-trained more guideline consistent (ie, screen less)
Zapka et al (2011)⁴⁶	CRC screening	Frequency of presenting >1 test option — Frequency of recommending FOBT + COL or COL alone compared to other tests or combinations	OR [95% CI]: Foreign trained^* US-trained: 0.5 [0.4, 0.7] — Foreign-trained ^* US-trained: 0.9 [0.6, 1.3]	US-trained less likely to discuss >1 test option (P = .0002) — Null (P = .4895)
Singal et al (2013)³⁹	CRC screening	Receipt of guideline-consistent CRC screening (FOBT, FS, barium enema, COL)	Percentages: US-trained: 58.2 Foreign-trained: 53.3	US-trained screen more (white patients, P < .001; black, P = .006; Hispanic, P = .033)

CBE, clinical breast exam; CI, confidence interval; COL, colonoscopy; CRC, colorectal cancer; FOBT, fecal occult blood test; FS, flexible sigmoidoscopy; OR, odds ratio; yrs, years.

Reference.

Board certification

Two studies found that board-certified physicians screen with mammography more often than those without such certification (Table 5);^40;45 another study found that non–board-certified physicians are more likely to under-screen (screen less often than guidelines specify) with mammography,⁵⁵ although a fourth study found no association.⁴² One study found that non–board-certified physicians over-screened with Pap tests,⁵⁵ and another determined that board-certified physicians were more guideline consistent than those without certification.³⁸

Table 5.

Summary of Findings from Articles That Analyzed the Impact of Physician Board Certification on Cancer Screening

Author	Screening test	Measure	Results	Overall effect (OR [95% CI])
Leach et al (2012)⁴²	Mammogram	Over-recommend: would recommend mammogram alone or in combination with CBE for a woman with a terminal comorbidity	Percentages: BC^* : 43.8 NBC: 54.5	Null (OR not reported)
Tan et al (2013)⁴⁰	Mammogram	Patients receiving mammogram	Percentage (listed as patient life expectancy <7 yrs; ≥7 yrs): NBC ^*: 29.9; 52.8 BC: 33.6; 58.1	BC screen more (1.09 [1.03, 1.15]; 1.13 [1.08–1.18])
Heflin et al (2006)⁵⁵	Mammogram Pap test	Over-screen: somewhat/very likely to offer mammogram to woman with life expectancy <5yr Under-screen: less likely to offer mammogram to woman with life expectancy >10 yrs Over-screen: somewhat/very likely to offer Pap test to woman with life expectancy <5 yr	Percentages (listed as over-screen with mammogram; under-screen with mammogram; over-screen with Pap test): BC: 46.3; 12.3; 23.0 NBC: 36.5; 22.9; 33.9	Over-screen with mammogram: null (P = .063) NBC more often under-screen with mammogram (P = .0061) NBC more often over-screen with Pap test (P = .017)
Yabroff et al (2009)³⁸	Pap test	Guideline-consistent response to clinical vignettes	OR [95% CI]: NBC^* BC: 2.02 [1.24, 3.27]	BC more guideline consistent
Pham et al (2005)⁴⁵	Mammogram CRC screening	Receipt of mammogram and CRC screening by Medicare beneficiaries age ≥65 yrs from US physician respondents to the 2000–2001 Community Tracking Study Physician Survey	Percentages (listed as mammogram, CRC screening): NBC^*: 36.5; 6.5 BC: 48.5; 9.5	BC screen more with mammogram (1.34 [1.04, 1.71]) and for CRC (1.27 [1.04, 1.57])
Cooper et al (1997)⁶⁵	CRC screening	Recommendation of screening with FOBT or FS for patients with comorbidities	Percentages (listed as no; mild; moderate; terminal comorbidities): FOBT: NBC: 90.7; 86.6; 81.0; 50.5 BC: 92.1; 85.5; 78.7; 32.3 FS: NBC: 33.3; 22.2; 13.9; 6.5 BC: 55.2; 30.7; 19.8; 4.2	NBC more likely to recommend screening with FOBT for patients with terminal comorbidities (P < .001) NBC less likely to recommend FS for patients with no (P < 0.001), mild (P < .05) or moderate (P < .05) comorbidities
Yabroff et al (2011)⁴⁴	CRC screening	All guideline-consistent responses to clinical vignettes	OR [95% CI]: NBC^* BC: 1.95 [1.12, 3.38]	BC more guideline consistent (ie, screen less)
Zapka et al (2011)⁴⁶	CRC screening	Frequency of recommending FOBT + COL or COL alone compared to other tests or combinations	OR [95% CI]: BC ^* NBC: 0.7 [0.5, 1.2]	Null (P = .2030)
Goodwin et al (2014)⁵³	PSA test	Patients receiving PSA	Percentages (listed as 2007; 2010): BC ^*: 33.6; 30.5 NBC: 29.6; 29.0	Null (0.92 [0.70, 1.22]; 0.94 [0.74, 1.20])
Klabunde et al (2012)⁶⁸	Chest radiograph LDCT	Recommendation of chest radiograph or LDCT for asymptomatic patient	OR [95% CI] (listed as chest radiograph; LDCT): BC ^* NBC: 0.8 [0.5, 1.3]; 0.9 [0.6, 1.5]	Null

BC, board-certified; CBE, clinical breast exam; CI, confidence interval; COL, colonoscopy; CRC, colorectal cancer; FOBT, fecal occult blood test; FS, flexible sigmoidoscopy; LDCT, low-dose computed tomography; NBC, non–board-certified; OR, odds ratio; PSA, prostate-specific antigen; yrs, years.

Reference.

In 2 studies, board-certified physicians were found to perform more colon cancer screening than other physicians,^45,65 but Yabroff et al found that board-certified physicians were more guideline consistent (less likely to over-screen) than others,⁴⁴ and Zapka et al found no association.⁴⁶ The one study of lung cancer screening found no association with board certification.⁶⁸

Specialty

Three studies found that obstetrician/gynecologists (OB/GYNs) perform more mammographic screening than internists and family practitioners (FPs) or general practitioners (GPs) (Supplementary Table S2).^36,42,71 Five studies also found that FPs screen less often than internists,^{36,40,45,49,54} but Leach et al found that internists screen less often than FPs.⁴² Four studies found no association with mammography.^54,55,60,72

Three studies found no association of specialty with gynecological cancer screening,^55,62,72 but 6 found that OB/GYNs screen more often FPs/GPs.^{36,37,43,54,61,63} All but 1 of these studies found that internists screen less often than FPs/GPs.⁶³ One found that internists screen more than FPs.⁶⁰ Two studies found that OB/GYNs were less guideline consistent than internists.^38,50

Armour et al found that internists screen more often for CRC than physicians in other specialties.⁷³ Two studies found that internists screen more often than OB/GYNs.^41,54 Four studies found that internists screen more often than FPs/GPs.^39,44,45,60 However, Zapka et al found that FPs were more likely to discuss more than 1 test option than internists,⁴⁶ and Richards et al found that GPs screen more often with FOBT than internists.⁵¹ Three additional studies found no association.^58,65,72

Specialty was not consistently associated with PSA testing. Four of 7 studies found no association.^20,52,53,74 Two of the remaining 3 suggested that FPs/GPs screened more often than internists,^67,75 but the third suggested that internists screened more often than FPs/GPs.⁷⁰ The one study of lung cancer screening found that internists screened more often with low-dose computed tomography (LDCT) than FPs/GPs.⁶⁸

Practice characteristics

Experience

None of the 3 studies of breast cancer screening found an association between years of experience and screening rates (Supplementary Table S3).^45,55,76 Most studies of gynecological cancer screening also found no association,^55,61,63,76 although 1 found a positive relationship.⁴³ Two of 4 studies found no association of CRC screening with physician experience.^45,73 One found a positive association of experience with the use of FOBT among patients with terminal comorbidities, but a stronger inverse relationship of experience with flexible sigmoidoscopy use in patients with no, mild, or moderate comorbidities.⁶⁵ Singal et al also found an inverse relationship of experience with any form of CRC screening.³⁹ For PSA testing, 2 of 3 studies found no association.^20,52 Edlefsen et al did find a positive relationship of experience with screening among male physicians, but no association among female physicians.⁶⁷ The one study of lung cancer screening found a positive relationship between experience and screening with LDCT or chest radiographs.⁶⁸

Volume of patients

Most studies found positive associations between the volume of patients seen and cancer screening practices (Supplementary Table S4). For breast cancer, one study found no association,⁷⁶ but another found that physicians with more patients screened more often with mammography for patients with a life expectancy of at least 7 years than physicians with fewer patients.⁴⁰ Similarly, physicians who saw fewer patients per week reported more guideline-consistent CRC screening practices than those who saw more patients.⁴⁴ One study found that physicians with more patients were more likely than those with fewer patients to have a policy of trying to talk patients into getting a PSA test,⁵² and another found a positive relationship between number of patients seen and PSA testing rates.⁵³ However, physicians with fewer patients were more likely to screen for lung cancer with chest radiography than physicians with greater patient volume.⁶⁸

Academic affiliation

Three of 4 cross-sectional studies found no association of academic affiliation with cancer screening practices (Supplementary Table S5).^38,42,44 The fourth found that non-affiliated physicians were less likely to recommend FOBT with colonoscopy or colonoscopy alone versus other screening types for CRC than physicians in academic settings, but the study did not explore overall testing frequency.⁴⁶

Discussion

This systematic review analyzed 42 studies exploring the associations of 10 physician characteristics with screening for breast, colorectal, cervical, prostate, and lung cancers. The aim was to examine the roles of those characteristics in discrepancies between screening frequency and the guidelines regarding specific cancer screening tests. The study team found that:

Female physicians, OB/GYNs or internists, and board-certified physicians were more likely than males and non-specialists to perform, to refer for, or to recommend mammographic screening.

Asian physicians were less likely and non–board-certified physicians, OB/GYNs, and FPs more likely to screen patients with Pap tests.

White physicians, internists, and those with more patients were more likely to screen patients for CRC than Asian or black physicians, other specialists, or physicians with smaller practices.

Black physicians and those with more patients were more likely to screen patients with PSA tests than physicians of other races.

Because only 1 study of lung cancer screening was found, the study team was unable to draw conclusions about relevant physician characteristics; the team suggests that future studies examine this topic.

Several of the aforementioned observations may reflect physician diligence about screening for patients with whom they personally identify.⁷⁷ Black physicians may screen more for prostate cancer because of their concern about the high prevalence of prostate cancer among black men.^20,52 Likewise, female physicians may screen more for breast cancer because they identify with their female patients and because female patients may be more likely to follow through with obtaining a mammogram when a female physician suggests it.⁵⁶

Differences in practices associated with physician specialty may have arisen because subspecialists are more confident than generalists in the guidelines developed by the governing organization of their particular subspecialty.^36,78 Leach et al suggest that OB/GYNs make greater use of mammography than other physicians because they are more often targets of medical malpractice suits.⁴² Internists may refer patients to subspecialists for preventive care more often than do FPs and GPs because they have more established subspecialist referral networks.⁴⁵

Specialty and sex were by far the most common characteristics studied, and the associations of screening with those variables were consistent across studies. Future investigators of screening practices might contribute to the literature by examining the association of screening practices with physician race/ethnicity, practice volume, and board certification.

The studies included were published between 1989 and 2015, a time frame during which cancer screening guidelines changed dramatically. Most of the studies were cross-sectional, and each presented data on rates of screening or consistency with the guidelines at its time of publication. The ACS has updated and publicized its guidelines for breast, cervical, colorectal, prostate, and lung cancer screening every few years since 2001.⁷⁹ Therefore, it would be beneficial for future studies to examine changes in screening practices in relation to physician characteristics before and after changes in those guidelines, as Goodwin et al did.⁵³

The study team did not perform meta-analyses because the 42 studies varied in design and outcomes. For example, some studies reported on the proportions of patients who had or reported having a given test, whereas others analyzed physician responses to hypothetical vignettes. Given the small sample (42 papers) and the diversity of approaches, meta-analysis would have necessitated a large number of calculations that would have been difficult to interpret meaningfully.

Several studies found only that physicians with a certain characteristic used certain screening tests more or less than others without that characteristic. The studies that analyzed the consistency of screening-related behavior with guidelines were more useful from a public health perspective, especially when they specified the direction of the inconsistencies.^38,44 A few studies also assessed physicians' beliefs about the efficacy of screening and potential influences, such as clinical evidence or patient insistence.^38,44 It also may be useful to consider how physician characteristics and screening-related behavior relate to knowledge of or agreement with screening guidelines, or to the potential economic benefits of screening.

This review had several limitations. Given the number of characteristics and screening tests that could be used as search terms, the study team may not have found all relevant studies, despite hand-searching bibliographies. In addition, two thirds of the studies reviewed were surveys of patients or physicians rather than chart or database reviews that documented actual practice; thus, the findings may have reflected reporting biases. Some physician surveys asked for opinions but did not relate the opinions to the guidelines. Patients' reports on their past medical or screening history also may have been subject to faulty recall. Most of the physician surveys (59%) used data from the American Medical Association Master File. If their purpose is to relate physician characteristics to the proportions of patients actually tested, future studies should be based on patient chart reviews or billing records.

Another possible source of bias was the overall agreement among authors about the criteria for article selection. Figure 1 lists those criteria; through them the study team may have inadvertently excluded some articles that would have provided a different slant on the determinants of physician screening behavior.

However, this systematic review highlights several physician characteristics that were correlated with cancer screening rates, namely physician sex, specialty, and race/ethnicity. Based on these results, members of the medicine and public health communities can better understand the discrepancies between recommendations and physician practices. As electronic health records and health communications systems become more sophisticated, it is increasingly possible to notify physicians when patients are due for a given test and when referrals for tests are inappropriate. We can then study the effects of such communications on both physician and patient adherence to the guidelines, given their individual characteristics.

Footnotes

Author Disclosure Statement

Dr. Neugut has consulted for Pfizer, Teva Pharmaceuticals, Eisai Pharmaceuticals, Otsuka Pharmaceuticals, and United Biosource Corporation, and is on the Medical Advisory Board of EHE, Intl. None of these associations is related to the topic of this study. The other authors have no conflicts of interest. Ms. MacLean was supported by a grant through the Brain Tumor Foundation.

Supplementary Material

Supplementary Table1

Supplementary Table2

Supplementary Table3

Supplementary Table4

Supplementary Table5

References

Stone

, Crawford

. To screen or nor to screen: the prostate cancer dilemma. Asian J Androl, 2015; 17:44–45.

Tabayoyong

, Abouassaly

. Prostate cancer screening and the associated controversy. Surg Clin North Am, 2015; 95:1023–1039.

Pox

CP.

Controversies in colorectal cancer screening. Digestion, 2014; 89:274–281.

Pivot

, Viguier

, Touboul

, et al. Breast cancer screening controversy: too much or not enough?. Eur J Cancer Prev, 2015; 24 suppl:S73–S76.

Schwartz

, Woloshin

, Fowler

Jr ., Welch

. Enthusiasm for cancer screening in the United States. JAMA, 2004; 291:71–78.

Sutkowi-Hemstreet

, Vu

, Harris

, Brewer

, Dolor

, Sheridan

. Adult patients' perspectives on the benefits and harms of overused screening tests: a qualitative study. J Gen Intern Med, 2015; 30:1618–1626.

Maurice

, Abouassaly

. Patient opinions on prostate cancer screening are swayed by the United States Preventative Services Task Force recommendations. Urology, 2014; 84:295–299.

Smith

, Manassaram-Baptiste

, Brooks

, et al. Cancer screening in the United States, 2015: a review of current American cancer society guidelines and current issues in cancer screening. CA Cancer J Clin, 2015; 65:30–54.

Kadiyala

Are U.S. cancer screening test patterns consistent with guideline recommendations with respect to the age of screening initiation?. BMC Health Serv Res, 2009; 9:185.

10.

Han

, Davis

, Weissfeld

, Dinse

. Generational risks for cancers not related to tobacco, screening, or treatment in the United States. Cancer, 2010; 116:940–948.

11.

Fairfield

, Gerstein

, Levin

, Stringfellow

, Wierman

, McNaughton-Collins

. Decisions about medication use and cancer screening across age groups in the United States. Patient Educ Couns, 2015; 98:338–343.

12.

Wools

, Dapper

, de Leeuw

. Colorectal cancer screening participation: a systematic review. Eur J Public Health, 2016; 26:158–168.

13.

Hoffman

, Elmore

, Fairfield

, Gerstein

, Levin

, Pignone

. Lack of shared decision making in cancer screening discussions: results from a national survey. Am J Prev Med, 2014; 47:251–259.

14.

Ritvo

, Myers

, Paszat

, Serenity

, Perez

, Rabeneck

. Gender differences in attitudes impeding colorectal cancer screening. BMC Public Health, 2013; 13:500.

15.

Gourlay

, Lewis

, Preisser

, Mitchell

, Sloane

. Perceptions of informed decision making about cancer screening in a diverse primary care population. Fam Med, 2010; 42:421–427.

16.

Peterson

, Murff

, Ness

, Dittus

. Colorectal cancer screening among men and women in the United States. J Womens Health (Larchmt), 2007; 16:57–65.

17.

Purc-Stephenson

, Gorey

. Lower adherence to screening mammography guidelines among ethnic minority women in America: a meta-analytic review. Prev Med, 2008; 46:479–488.

18.

Katz

, Claudio

, Kressin

, Green

, Wang

, Russell

. Willingness to participate in cancer screenings: blacks vs whites vs Puerto Rican Hispanics. Cancer Control, 2008; 15:334–343.

19.

Pruitt

, Shim

, Mullen

, Vernon

, Amick BC 3rd. Association of area socioeconomic status and breast, cervical, and colorectal cancer screening: a systematic review. Cancer Epidemiol Biomarkers Prev, 2009; 18:2579–2599.

20.

Pollack

, Platz

, Bhavsar

, et al. Primary care providers' perspectives on discontinuing prostate cancer screening. Cancer, 2012; 118:5518–5524.

21.

Felsen

, Piasecki

, Ferrante

, Ohman-Strickland

, Crabtree

. Colorectal cancer screening among primary care patients: does risk affect screening behavior?. J Community Health, 2011; 36:605–611.

22.

Schroy

3rd , Caron

, Sherman

, Heeren

, Battaglia

. Risk assessment and clinical decision making for colorectal cancer screening. Health Expect, 2015; 18:1327–1338.

23.

Marcus

, Freedman

, Khoury

. Targeted cancer screening in average-risk individuals. Am J Prev Med, 2015; 49:765–771.

24.

Memon

, Kanwal

, Sami

, Larik

, Farooq

. Risk of breast cancer among young women and importance of early screening. Asian Pac J Cancer Prev, 2015; 16:7485–7489.

25.

Liu

, O'Malley

, Mori

, et al. The association of type and number of chronic diseases with breast, cervical, and colorectal cancer screening. J Am Board Fam Med, 2014; 27:669–681.

26.

Lansdorp-Vogelaar

, Gulati

, Mariotto

, et al. Personalizing age of cancer screening cessation based on comorbid conditions: model estimates of harms and benefits. Ann Intern Med, 2014; 161:104–112.

27.

Lukin

, Jandorf

, Dhulkifl

, et al. Effect of comorbid conditions on adherence to colorectal cancer screening. J Cancer Educ, 2012; 27:269–276.

28.

Carney

, O'Malley

, Buckley

, et al. Influence of health insurance coverage on breast, cervical, and colorectal cancer screening in rural primary care settings. Cancer, 2012; 118:6217–6225.

29.

Cowburn

, Carlson

, Lapidus

, DeVoe

. The association between insurance status and cervical cancer screening in community health centers: exploring the potential of electronic health records for population-level surveillance, 2008–2010. Prev Chronic Dis, 2013; 10:E173.

30.

Swan

, Breen

, Graubard

, et al. Data and trends in cancer screening in the United States: results from the 2005 National Health Interview Survey. Cancer, 2010; 116:4872–4881.

31.

Messina

, Lane

, Colson

. Colorectal cancer screening among users of county health centers and users of private physician practices. Public Health Rep, 2009; 124:568–578.

32.

Debarros

, Steele

. Colorectal cancer screening in an equal access healthcare system. J Cancer, 2013; 4:270–280.

33.

Committee on Health Care for Underserved Women. Committee Opinion No. 624: Cervical cancer screening in low-resource settings. Obstet Gynecol, 2015; 125:526–528.

34.

Walsh

, Posner

, Perez-Stable

. Colon cancer screening in the ambulatory setting. Prev Med, 2002; 35:209–218.

35.

Campos

, Castle

, Wright

Jr ., Kim

. Cervical cancer screening in low-resource settings: a cost-effectiveness framework for valuing tradeoffs between test performance and program coverage. Int J Cancer, 2015; 137:2208–2219.

36.

Haggstrom

, Phillips

, Liang

, Haas

, Tye

, Kerlikowske

. Variation in screening mammography and Papanicolaou smear by primary care physician specialty and gatekeeper plan (United States). Cancer Causes Control, 2004; 15:883–892.

37.

Price

RA.

Association between physician specialty and uptake of new medical technologies: HPV tests in Florida Medicaid. J Gen Intern Med, 2010; 25:1178–1185.

38.

Yabroff

, Saraiya

, Meissner

, et al. Specialty differences in primary care physician reports of papanicolaou test screening practices: a national survey, 2006 to 2007. Ann Intern Med, 2009; 151:602–611.

39.

Singal

, Lin

, Kuo

, Riall

, Goodwin

. Primary care physicians and disparities in colorectal cancer screening in the elderly. Health Serv Res, 2013; 48:95–113.

40.

Tan

, Kuo

, Elting

, Goodwin

. Refining physician quality indicators for screening mammography in older women: distinguishing appropriate use from overuse. J Am Geriatr Soc, 2013; 61:380–387.

41.

Zapka

, Klabunde

, Taplin

, Yuan

, Ransohoff

, Kobrin

. Screening colonoscopy in the US: attitudes and practices of primary care physicians. J Gen Intern Med, 2012; 27:1150–1158.

42.

Leach

, Klabunde

, Alfano

, Smith

, Rowland

. Physician over-recommendation of mammography for terminally ill women. Cancer, 2012; 118:27–37.

43.

Stormo

, Cooper

, Hawkins

, Saraiya

. Physician characteristics and beliefs associated with use of pelvic examinations in asymptomatic women. Prev Med, 2012; 54:415–421.

44.

Yabroff

, Klabunde

, Yuan

, et al. Are physicians' recommendations for colorectal cancer screening guideline-consistent?. J Gen Intern Med 2011:26:177–184.

45.

Pham

, Schrag

, Hargraves

, Bach

. Delivery of preventive services to older adults by primary care physicians. JAMA, 2005; 294:473–481.

46.

Zapka

, Klabunde

, Arora

, Yuan

, Smith

, Kobrin

. Physicians' colorectal cancer screening discussion and recommendation patterns. Cancer Epidemiol Biomarkers Prev, 2011; 20:509–521.

47.

Miller

, Baldwin

, Matthews

, et al. Physicians' beliefs about effectiveness of cancer screening tests: a national survey of family physicians, general internists, and obstetrician-gynecologists. Prev Med, 2014; 69:37–42.

48.

Brownson

, Davis

, Simms

, Kern

, Harmon

. Cancer control knowledge and priorities among primary care physicians. J Cancer Educ, 1993; 8:35–41.

49.

Lane

, Zapka

, Breen

, Messina

, Fotheringham

. A systems model of clinical preventive care: the case of breast cancer screening among older women. For the NCI Breast Cancer Screening Consortium. Prev Med, 2000; 31:481–493.

50.

Bruder

, Downes

, Malo

, Giuliano

, Salmon

, Vadaparampil

. Physicians' intentions to change pap smear frequency following human papillomavirus vaccination. J Pediatr Adolesc Gynecol, 2012; 25:384–389.

51.

Richards

, Klabunde

, O'Malley

. Physicians' recommendations for colon cancer screening in women. Too much of a good thing?. Am J Prev Med, 1998; 15:246–249.

52.

Hall

, Taylor

, Ross

, Richardson

, Richards

, Rim

. Discussions about prostate cancer screening between U.S. primary care physicians and their patients. J Gen Intern Med, 2011; 26:1098–1104.

53.

Goodwin

, Jaramillo

, Yang

, Kuo

, Tan

. Is anyone listening? Variation in PSA screening among providers for men 75+ before and after United States Preventive Services Task Force recommendations against it: a retrospective cohort study. PLoS One, 2014; 9:e107352.

54.

Weisman

, Celentano

, Teitelbaum

, Klassen

. Cancer screening services for the elderly. Public Health Rep, 1989; 104:209–214.

55.

Heflin

, Pollak

, Kuchibhatla

, Branch

, Oddone

. The impact of health status on physicians' intentions to offer cancer screening to older women. J Gerontol A Biol Sci Med Sci, 2006; 61:844–850.

56.

Lurie

, Slater

, McGovern

, Ekstrum

, Quam

, Margolis

. Preventive care for women. Does the sex of the physician matter?. N Engl J Med, 1993; 329:478–482.

57.

Tishler

, McCarthy

, Rind

, Hamel

. Breast cancer screening for older women in a primary care practice. J Am Geriatr Soc, 2000; 48:961–966.

58.

Liang

, Wang

, Chen

, Mandelblatt

. Language use and the receipt of cancer screening recommendations by immigrant Chinese American women. J Womens Health (Larchmt), 2009; 18:201–207.

59.

Lurie

, Margolis

, McGovern

, Mink

, Slater

. Why do patients of female physicians have higher rates of breast and cervical cancer screening?. J Gen Intern Med, 1997; 12:34–43.

60.

Thompson

, Gomez

, Chan

, et al. Patient and provider characteristics associated with colorectal, breast, and cervical cancer screening among Asian Americans. Cancer Epidemiol Biomarkers Prev, 2014; 23:2208–2217.

61.

Almeida

, Rodriguez

, Skootsky

, Pregler

, Steers

, Wenger

. Cervical cancer screening overuse and underuse: patient and physician factors. Am J Manag Care, 2013; 19:482–489.

62.

Arredondo

, Pollak

, Costanzo

, McNeilly

, Myers

. Primary care residents' characteristics and motives for providing differential medical treatment of cervical cancer screening. J Natl Med Assoc, 2003; 95:576–584.

63.

Saraiya

, Berkowitz

, Yabroff

, Wideroff

, Kobrin

, Benard

. Cervical cancer screening with both human papillomavirus and Papanicolaou testing vs Papanicolaou testing alone: what screening intervals are physicians recommending?. Arch Intern Med, 2010; 170:977–985.

64.

Taylor

, Schwartz

, Yasui

, et al. Pap testing among Vietnamese women: health care system and physician factors. J Community Health, 2004; 29:437–450.

65.

Cooper

, Fortinsky

, Hapke

, Landefeld

. Primary care physician recommendations for colorectal cancer screening. Patient and practitioner factors. Arch Intern Med, 1997; 157:1946–1950.

66.

Shokar

, Nguyen-Oghalai

, Wu

. Factors associated with a physician's recommendation for colorectal cancer screening in a diverse population. Fam Med, 2009; 41:427–433.

67.

Edlefsen

, Mandelson

, McIntosh

, Andersen

, Wagner

, Urban

. Prostate-specific antigen for prostate cancer screening. Do physician characteristics affect its use?. Am J Prev Med, 1999; 17:87–90.

68.

Klabunde

, Marcus

, Han

, et al. Lung cancer screening practices of primary care physicians: results from a national survey. Ann Fam Med, 2012; 10:102–110.

69.

Lew

, Moskowitz

, Ngo

, et al. Effect of provider status on preventive screening among Korean-American women in Alameda County, California. Prev Med, 2003; 36:141–149.

70.

Wallner

, Frencher

, Hsu

, et al. Prostate cancer screening trends in a large, integrated health care system. Perm J, 2012; 16:4–9.

71.

Bhosle

, Samuel

, Vosuri

, Paskett

, Balkrishnan

. Physician and patient characteristics associated with outpatient breast cancer screening recommendations in the United States: analysis of the National Ambulatory Medical Care Survey Data 1996–2004. Breast Cancer Res Treat, 2007; 103:53–59.

72.

DuBard

, Schmid

, Yow

, Rogers

, Lawrence

. Recommendation for and receipt of cancer screenings among medicaid recipients 50 years and older. Arch Intern Med, 2008; 168:2014–2021.

73.

Armour

, Friedman

, Pitts

, Wike

, Alley

, Etchason

. The influence of year-end bonuses on colorectal cancer screening. Am J Manag Care, 2004; 10:617–624.

74.

Liang

, Phillips

, Tye

, Haas

, Sakowski

. Does patient cost sharing matter? Its impact on recommended versus controversial cancer screening services. Am J Manag Care, 2004; 10(2 Pt 1):99–107.

75.

Tasian

, Cooperberg

, Potter

, et al. PSA screening: determinants of primary-care physician practice patterns. Prostate Cancer Prostatic Dis, 2012; 15:189–194.

76.

Parkerton

, Smith

, Straley

. Primary care practice coordination versus physician continuity. Fam Med, 2004; 36:15–21.

77.

Street

Jr ., O'Malley

, Cooper

, Haidet

Understanding concordance in patient-physician relationships: personal and ethnic dimensions of shared identity. Ann Fam Med, 2008; 6:198–205.

78.

Tunis

, Hayward

, Wilson

, et al. Internists' attitudes about clinical practice guidelines. Ann Intern Med, 1994; 120:956–963.

79.

Smith

, Andrews

, Brooks

, et al. Cancer screening in the United States, 2017: a review of current American Cancer Society guidelines and current issues in cancer screening. CA Cancer J Clin, 2017; 67:100–121.

80.

Tyler

Jr ., Snyder

. Cancer risk assessment: examining the family physician's role. J Am Board Fam Med, 2006; 19:468–477.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB