Refer to USPSTF Procedure Manual for interpretation of the analytic framework.16
KQ indicates key question.
aReasons for exclusion: Relevance: Study aim not relevant; Design: Study did not use an included design; Setting: Study was not conducted in a country relevant to US practice; Population: Study was not conducted in an unselected population; Outcomes: Study did not have relevant outcomes or had incomplete outcomes; Intervention: Study used an excluded intervention/screening approach; Poor quality: Study did not meet criteria for fair or good quality; Condition: Study examined an excluded condition (cervical cancer, gonorrhea, or chlamydia).
bEight studies (in 10 articles) were included for both KQ2 and KQ3.
eMethods. Literature Search Strategies for Primary Literature
eTable 1. Inclusion and Exclusion Criteria
eTable 2. Quality Assessment Criteria
Customize your JAMA Network experience by selecting one or more topics from the list below.
Guirguis-Blake JM, Henderson JT, Perdue LA. Periodic Screening Pelvic ExaminationEvidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2017;317(9):954–966. doi:10.1001/jama.2016.12819
Copyright 2017 American Medical Association. All Rights Reserved.
Recent changes in the periodicity of cervical cancer screening have led to questions about the role of screening pelvic examinations among asymptomatic women.
To systematically review literature on health benefits, accuracy, and harms of the screening pelvic examination for gynecologic conditions for the US Preventive Services Task Force (USPSTF).
MEDLINE, PubMed, and Cochrane Central Register of Controlled Trials for relevant English-language studies published through January 13, 2016, with surveillance through August 3, 2016.
Two reviewers independently screened abstracts and studies. The search yielded 8678 unique citations; 316 full-text articles were reviewed, and 9 studies including 27 630 patients met inclusion criteria.
Data Extraction and Synthesis
Two reviewers rated study quality using USPSTF criteria.
Main Outcomes and Measures
Morbidity; mortality; diagnostic accuracy for any gynecologic cancer or condition except cervical cancer, gonorrhea, and chlamydia, which are covered by other USPSTF screening recommendations; harms (false-positive rates, false-negative rates, surgery rates).
No trials examined the effectiveness of the pelvic examination in reducing all-cause mortality, reducing cancer- and disease-specific morbidity and mortality, or improving quality of life. Eight studies reported accuracy for the screening pelvic examination: ovarian cancer (4 studies; n = 26 432), bacterial vaginosis (2 studies; n = 930), trichomoniasis (1 study; n = 779), and genital herpes (1 study; n = 779). In the 4 ovarian cancer screening studies, low prevalence of ovarian cancer consistently resulted in low positive predictive values (PPVs) and false-positive rates, with a lack of precision in accuracy estimates (sensitivity range, 0%-100%; specificity range, 91%-99%; PPV range, 0%-3.6%; negative predictive value [NPV] range, ≥99%). Each diagnostic accuracy study for bacterial vaginosis, trichomoniasis, and genital herpes was performed in a high-prevalence population with substantial proportions of symptomatic patients and reported accuracy characteristics for individual physical examination findings (bacterial vaginosis, homogeneous discharge: sensitivity range, 69%-79%; specificity range, 54%-97%; PPV range, 52%-95%; NPV range, 79%-80%; herpes simplex virus, vulvar ulcerations: sensitivity, 20%; specificity, 98%; PPV, 88%; NPV, 57%; trichomoniasis, colpitis macularis: sensitivity, 2%; specificity, 100%; PPV, 100%; NPV, 85%). Surgery rates resulting from an abnormal screening pelvic examination for ovarian cancer ranged from 5% to 36% at 1 year, with the largest study reporting an 11% surgery rate and 1% complication rate within 1 year of a screening pelvic examination with abnormal findings.
Conclusions and Relevance
No direct evidence was identified for overall benefits and harms of the pelvic examination as a 1-time or periodic screening test. Limited evidence was identified regarding the diagnostic accuracy and harms of routine screening pelvic examinations in asymptomatic primary care populations.
In 2012, 44.2 million pelvic examinations were performed in US outpatient visits,1 and in a 2008-2009 survey of physicians, 69.1% agreed that the pelvic examination is an effective screening test for ovarian cancer.2 Multiple malignant and benign gynecologic conditions can be detected with pelvic examination, and the value of early detection among asymptomatic women depends on natural history, prevalence, disease morbidity, and early treatment effectiveness.3,4 For some gynecologic conditions, alternative effective screening approaches are recommended (eg, nucleic acid amplification test for gonorrhea and chlamydia,5 Papanicolaou smear and human papillomavirus testing for cervical cancer6). As recommended intervals for cervical cancer screening have been lengthened,6-8 the health benefits of annual pelvic examinations among asymptomatic women have increasingly been questioned.9-12 Current professional organizations’ guidelines for the administration of the screening pelvic examination are based on limited evidence and expert opinion, and these guidelines vary widely.13-15 The current evidence review was undertaken to inform US Preventive Services Task Force (USPSTF) deliberations on whether nonpregnant women without gynecologic symptoms would obtain net health benefits from periodic screening pelvic examinations.
In this evidence review, the pelvic examination was defined as visual inspection of the external genitalia; a speculum examination of the vagina and cervix; bimanual examination of the adnexa, uterus, and cervix; and may include the rectovaginal examination.13 This review addresses the benefits and harms of screening with the pelvic examination for benign and malignant gynecologic conditions as well as the diagnostic accuracy of the pelvic examination in detecting individual benign and malignant gynecologic conditions. The analytic framework and key questions that guided the review are shown in Figure 1. Detailed methods are available in the full evidence report at https://www.uspreventiveservicestaskforce.org/Page/Document/final-evidence-review140/gynecological-conditions-screening-with-the-pelvic-examination.
Although the pelvic examination is commonly performed in adolescent and pregnant women, these populations were specifically excluded from the scope of this review. The purpose of conducting the pelvic examination in unselected nonpregnant adult women often differs from that in special populations of adolescents (eg, Tanner staging, congenital abnormality case-finding) or pregnant women (eg, assessment of pregnancy dating, pelvic outlet adequacy, cervical dilation). Likewise, screening for congenital gynecological conditions was excluded because this review focuses on routine periodic screening, and many of the congenital conditions would be detected at the symptomatic stage, during pregnancy, at infertility workup, or incidentally during cervical cancer screening.
In addition, the USPSTF previously determined that there is good evidence for primary screening approaches for cervical cancer (such as Papanicolaou and human papillomavirus cotesting) and for gonorrhea and chlamydia (such as nucleic acid amplification testing). Since the pelvic examination alone is less accurate than these screening approaches, these conditions were not included in this review.
MEDLINE, PubMed (publisher-supplied references only), and the Cochrane Central Register of Controlled Trials were searched to locate primary studies informing the key questions (eMethods in the Supplement) and published from the earliest date indexed (1946 for MEDLINE) through January 13, 2016. The database searches were supplemented with experts’ suggestions and reference lists from all other recent systematic reviews.17,18 ClinicalTrials.gov and the World Health Organization International Clinical Trials Registry Platform were searched for ongoing trials. The National Cancer Institute provided previously unpublished data on the subset of randomized women from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial who had undergone bimanual palpation of the ovaries and adnexa as well as rectovaginal examination (Paul Pinsky, PhD, National Cancer Institute, written communication, May 2, 2016); the 5-year follow-up results were subsequently published.19 Since January 2016, ongoing surveillance was conducted through article alerts and targeted searches of high-impact journals to identify major studies published in the interim that may affect the conclusions or understanding of the evidence and therefore the related USPSTF recommendation. The last surveillance was conducted on August 3, 2016, and identified no new studies.
Two investigators independently reviewed 8678 titles and abstracts by using an online platform (abstrackr20) and 316 articles (Figure 2) with specified inclusion criteria (eTable 1 in the Supplement). Discrepancies were resolved through consensus and consultation with a third investigator. Articles that did not meet inclusion criteria or those rated as poor quality were excluded; criteria for establishing study quality are noted in eTable 2 in the Supplement. To avoid missing studies using the pelvic examination as a secondary screening test (eg, ovarian cancer screening studies using cancer antigen 125 [CA-125] measurement and ultrasound technology that also included a pelvic examination component), reviewers were more inclusive during the review of abstracts and titles. As a result, many studies were excluded at the full-text review.
Eligible studies included unselected adult females who were not symptomatic or pregnant and were conducted in developed countries, as defined by “very high” development according to the 2014 United Nations Human Development Index.21 Studies conducted solely in symptomatic populations were excluded.
Any study that examined the relationship between pelvic examination and all-cause mortality, cancer- or disease-specific morbidity or mortality, or quality of life was eligible for inclusion. In addition, studies examining the screening accuracy of the pelvic examination in a single encounter or as a periodic program of screening were eligible.
One reviewer extracted study-level data into standardized evidence tables, and a second checked for accuracy. At least 2 reviewers critically appraised included studies using the Newcastle Ottawa Scales22 for cohort and case-control studies and Quality Assessment of Diagnostic Accuracy Studies I and II23,24 for studies of diagnostic accuracy adapted to align with the USPSTF’s design-specific quality criteria16 (eTable 2 in the Supplement). Disagreements in quality rating were resolved by consensus or consultation with a third independent reviewer. Included studies were limited to those published in English that were rated as good or fair quality using USPSTF quality-rating standards.16
Results were qualitatively synthesized by key question (Figure 1). For all of the studies of diagnostic accuracy, sensitivity and specificity were calculated in Stata version 13.1 (StataCorp), using Jeffrey confidence intervals from 2 × 2 tables constructed from data reported in the primary studies. In many cases the data presented differ slightly from the data in the published article because of these calculations. For diagnostic accuracy studies, in addition to the standard test performance characteristics (sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]), the following outcomes were calculated: condition prevalence in the study population, percentage of patients screening positive, false-positive rate, and false-negative rate. Since there were a limited number of studies for each condition, no pooled analyses were conducted.
Nine studies met the inclusion criteria for this systematic review (Figure 2). The results are organized by key question. Eight of the 9 included studies reported outcomes for both key question 2 (accuracy) and key question 3 (harms). A single additional study on key question 3 (harms) was also included.
Key Question 1. What is the direct evidence for the effectiveness of the pelvic examination in (a) reducing all-cause mortality, (b) reducing cancer- and disease-specific morbidity and mortality, and (c) improving quality of life?
There was no direct evidence comparing the effectiveness of pelvic examination screening and no screening on patient health outcomes.
Key Question 2. What are the test performance characteristics of the pelvic examination (sensitivity, specificity, PPV, and NPV) in screening for gynecologic cancers and other gynecologic conditions?
Studies examining the diagnostic screening accuracy of the pelvic examination were identified for 4 gynecologic conditions: ovarian cancer, bacterial vaginosis, trichomoniasis, and genital herpes. In total, 4 studies examined the accuracy of the pelvic examination to detect ovarian cancer in asymptomatic populations, while 1 to 2 studies in high-prevalence populations with substantial proportions of symptomatic patients were identified for each of the infectious diseases (bacterial vaginosis, trichomoniasis, and genital herpes).
Data from 1 good-quality study and 3 fair-quality studies (n = 26 432) were identified to estimate the screening accuracy of pelvic examination for ovarian cancer detection.25-27 The PLCO trial (n = 20 872), a large multicenter US randomized clinical trial, recruited average-risk women aged 55 to 74 years from the community to test the benefits and harms of ovarian cancer screening using a combination of 3 modalities: measurement of CA-125 levels, ultrasound, and ovarian palpation/rectovaginal examination.28 Ovarian palpation was stopped after the first 5 years of the trial because no ovarian cancers were detected solely with this test; accuracy information and clinical outcomes resulting from ovarian palpation were not included in published results from the trial. The unpublished results were obtained from National Cancer Institute PLCO investigators. The other 3 were prospective diagnostic accuracy studies (n = 5560) conducted in Greece, Australia, and the United Kingdom; they primarily recruited average-risk women aged 40 to 45 years and older from the community (Table 1).
Mean or median age ranged from 51 to 63 years. In the 2 studies reporting menopausal status, 43%27 and 65%26 of participants were postmenopausal. In the other 2 studies, all or nearly all were postmenopausal.25,28 None of the studies excluded women with a family history of ovarian cancer, and 1 actively recruited younger women with family history.27 One study defined the test as bimanual ovarian palpation plus rectovaginal examination,28 another clearly defined the index pelvic examination test as bimanual and speculum examination,27 while the other 2 defined the index test only as a “pelvic exam”26 or “vaginal exam.”25 Two studies specified that experienced gynecologists or examiners performed the examination,26,28 and in the other studies a single examiner27 or 1 of 2 examiners25 performed the examinations.
A positive index test finding was a palpable pelvic or adnexal mass of any size. The PLCO trial used a reference standard for ovarian cancer diagnoses captured in medical records and participant questionnaires; investigators reported cancer incidence at 1 year and up to 5 years for the purposes of this analysis.19,28,29 The other 3 studies used a reference standard of incident ovarian cancer reported in a 1-year patient questionnaire and measurement of CA-125 levels with or without ultrasound.25-27 Abnormal ultrasound findings or CA-125 levels were variably defined in the different studies.
Ovarian cancer prevalence was reported as 0.1% in 3 studies25,26 and 0.04% in 1 study27; the longer follow-up from the PLCO trial (5 years) yielded 0.3% prevalence of ovarian cancer (Table 2). The comparable 1-year follow-up results for the proportion of screened patients with a positive pelvic examination finding ranged from 1.2% to 8.7%. Sensitivity was reported as 100.0% in 2 of the studies25,26 (1-2 ovarian cancer cases were palpable on pelvic examination) and as 0% (95% CI, 0%-85.3%) in the study in which the single case of ovarian cancer was not detected with the pelvic examination.27 The PLCO trial reported a sensitivity of 4.3% (95% CI, 0.5%-18.6%) and 2.8% (95% CI, 0.6%-8.6%) from the first screening examination at 1-year follow-up (23 cases) and up to 5 years’ follow-up (72 cases). Specificity ranged from 91.4% (95% CI, 90.1%-92.6%) to 98.8% (95% CI, 98.7%-99.0%). Calculated PPVs ranged from 0% (95% CI, 0%-6.0%) to 3.6% (95% CI, 0.4%-15.5%), and NPV was 99% or greater for all studies.19,25-27 In all studies except the PLCO trial, accuracy estimates had wide confidence intervals because of the very low event rate.
No screening studies were conducted solely in asymptomatic primary care populations. Two fair-quality US studies (n = 930) with large proportions of symptomatic patients assessed the accuracy of different approaches to diagnosing bacterial vaginosis, including findings visible on the pelvic examination.30,31 Gutman et al (n = 269)30 recruited any woman undergoing a “routine pelvic examination” from a hospital-based primary care clinic, colposcopy clinic, or research clinic, whereas Eschenbach et al (n = 661)31 recruited nonpregnant women aged 16 to 50 years from a sexually transmitted infection (STI) clinic undergoing a “standardized pelvic examination”(Table 3). Neither study had a primary aim of estimating the accuracy of the pelvic examination; instead, they explored different clinical signs and diagnostic criteria for bacterial vaginosis measured against a gold standard (Nugent criteria30 or criteria based on pH; Gram stain microscopy31,34).
The patients in the study by Gutman et al30 had a mean age of 24.1 years, with 38% of patients being white, 30% black, and 27% Hispanic; the study by Eschenbach et al31 did not report patient characteristics. In the study by Gutman et al,30 33% of patients were symptomatic, while the study by Eschenbach et al31 reported that 59% of participants presented with some pelvic or abdominal symptom as a chief presenting symptom. Both studies reported a high prevalence of bacterial vaginosis (39% and 47%).30,31 Gutman et al30 reported the sensitivity and specificity of thin, homogeneous discharge as 78.8% (95% CI, 70.3%-85.8%) and 53.9% (95% CI, 46.3%-61.4%), respectively (PPV, 51.9% [95% CI, 44.1%-59.6%] and NPV, 80.2% [95% CI, 72.0%-86.8%]) (Table 4). Eschenbach et al31 reported the sensitivity and specificity of homogeneous discharge as 69.2% (95% CI, 63.4%-74.5%) and 97.2% (95% CI, 94.9%-98.6%), respectively (PPV, 95.3% [95% CI, 91.7%-97.7%]; NPV, 79.0% [95% CI, 74.8%-82.8%]).
No screening studies were conducted solely in asymptomatic primary care populations. One fair-quality trial by Koutsky et al (n = 779)32 provided data on the accuracy of specific pelvic examination findings in detecting this condition. The study recruited nonpregnant women aged 16 to 50 years from the same population seen in the STI clinic for the bacterial vaginosis study by Eschenbach et al31 and for the trichomoniasis study by Wølner-Hansson et al33 (Table 3). All pelvic examinations were performed by 1 “women’s health care specialist.” Mean age was 24 years, and 70% of participants were white. Almost all patients were sexually active (98%). Condoms were used as the primary method of contraception by 7%, and 33% did not use any contraception. Of the participants, 10% were symptomatic. The index test was a “genital examination” (colposcopy findings were not considered for this review). A positive pelvic examination result was defined as clinician-detected lesions, and only vulvar ulcerations and tender inguinal nodes were used in this review for accuracy calculations. All patients received the reference test, which included cultures from urine, cervical swabs, anal swabs, and any lesion swabs (all herpes simplex virus [HSV] isolates were confirmed and typed by direct immunofluorescence with use of mouse monoclonal antibodies), as well as serum testing for HSV-1 or HSV-2 antibodies using Western blotting.
Nearly half (48%) of all study participants were diagnosed with genital herpes at some stage of the disease: 6% were diagnosed at the first episode, 5% had symptomatic recurrence, 2% had asymptomatic shedding, and 35% had latent HSV-2 infection (defined as HSV-2 antibodies present without signs or symptoms). Of those with any stage of genital herpes, 78% had latent disease. The presence of vulvar ulcerations had a sensitivity of 19.6% (95% CI, 15.8%-23.9%) and specificity of 97.5% (95% CI, 95.7%-98.7%) in detecting genital HSV at any stage (PPV, 88.0% [95% CI, 79.7%-93.6%]; NPV, 57.0% [95% CI, 53.3%-60.7%]) (Table 4). Similarly, the presence of tender inguinal lymphadenopathy had a sensitivity of 14.2% (95% CI, 11.0%-18.1%) and specificity of 97.1% (95% CI, 95.1%-98.4%) (PPV, 81.5% [95% CI, 70.8%-89.5%]; NPV, 55.3% [95% CI, 51.7%-58.9%]).32
There were no screening studies conducted solely in asymptomatic primary care populations. One fair-quality trial by Wølner-Hanssen et al (n = 779)33 was set in the same STI clinic as the trials by Eschenbach et al31 and Koutsky et al,32 with a high proportion of symptomatic patients aimed to analyze the clinical manifestations of trichomoniasis and determined the accuracy of specific clinical findings on pelvic examination in detecting trichomoniasis (Table 3). An abnormal pelvic examination finding was defined as clinician-reported moderate to markedly increased vaginal discharge compared with that seen in patients without genital infections. The reference test for identifying trichomoniasis was culture. At least half of the patients reported symptoms of yellow discharge (23%), abnormal vaginal odor (36%), or vulvar itching (51%). The prevalence of culture-confirmed trichomoniasis was 15%. For the most specific clinical sign, colpitis macularis or “strawberry cervix” (detected grossly, without a colposcope), sensitivity was 1.7% (95% CI, 0.4%-5.3%) and specificity was 100.0% (95% CI, 99.6%-100.0%) (PPV, 100.0% [95% CI, 33.3%-100.0%]; NPV, 85.1% [95% CI, 82.4%-87.4%]) (Table 4). For other individual clinical findings, sensitivity ranged from 8.0% (95% CI, 3.8%-14.5%) to 59.2% (95% CI, 49.3%-68.5%) and specificity from 76.1% (72.4%-79.5%) to 99.1% (98.1%-99.7%); PPVs ranged from 30.2% (95% CI, 24.0%-37.0%) to 61.5% (35.0%-83.5%) and NPVs from 86.3% (83.5%-88.7%) to 91.4% (95% CI, 88.6%-93.7%) (Table 4).
Key Question 3. What are the adverse effects of screening by pelvic examination?
All accuracy studies (8 studies) were included for harms (false-positive rates and resulting diagnostic workup), and 1 additional study was included for other harms.
Additional imaging and unnecessary surgical intervention are potential harms of pelvic examination screening for ovarian cancer. The prevalence of laparoscopy or laparotomy for those with abnormal findings on pelvic examination ranged from 5% to 36%.19,25-27 In the Greek study,26 17% of women with abnormal pelvic examination results underwent surgery because of the results. Pathology findings revealed 2 cases of ovarian cancer (1 metastatic, 1 stage Ia serous cystadenocarcinoma), 4 serous cystadenomas, 3 mucinous cystadenomas, 5 endometroid cysts of the ovary, 12 benign cysts, and 3 normal pathology results.26 In the Australian study,27 2 women (5%) with abnormal findings on pelvic examination had surgery. The operations revealed that 1 patient had a fibroid uterus and 1 a normal (negative) result; the single case of ovarian cancer was not detected on pelvic examination.27 In the UK study that recruited solely postmenopausal women,25 36% of women with an abnormal pelvic examination result underwent surgery based on the pelvic examination result: 1 woman had ovarian cancer, and 9 had benign conditions.25 In the PLCO trial, the surgery rate occurring subsequent to an ovarian palpation examination with abnormal findings was 11% over up to 4 rounds of screening (mean, 2.4 ovarian pelvic palpation examinations per woman), with a complication rate of 1%; additional ultrasounds, CA-125 measurements, pelvic examinations, and computed tomography scans occurring subsequent to an ovarian palpation examination with abnormal findings are reported in Table 5.
False-positive and false-negative rates for thin homogeneous discharge were 46.1% (95% CI, 38.6%-53.7%) and 21.2% (95% CI, 14.2%-29.7%), respectively.30 Eschenbach et al31 reported that false-positive and false-negative rates for homogeneous discharge were 2.8% (95% CI, 1.4%-5.1%) and 30.8% (95% CI, 25.5%-36.6%), respectively (Table 4).
False-positive and false-negative rates for vulvar ulcerations were 2.5% (95% CI, 1.3%-4.3%) and 80.4% (95% CI, 76.1%-84.2%), respectively, for any stage of genital herpes. False-positive and false-negative rates for tender inguinal lymphadenopathy were 2.9% (95% CI, 1.6%-4.9%) and 85.8% (95% CI, 81.9%-89.0%), respectively, for any stage of genital herpes (Table 4).32
False-positive and false-negative rates for colpitis macularis were 0% (95% CI, 0%-0.4%) and 98.3% (95% CI, 94.7%-99.6%), respectively (Table 4).33 False-positive and false-negative rates for purulent discharge were 23.9% (95% CI, 20.5%-27.6%) and 40.8% (95% CI, 31.5%-50.7%), respectively.
Beyond the downstream diagnostic workup and surgical procedures resulting from the pelvic examination for ovarian cancer screening and the false-positive rates and missed cases from accuracy studies, 1 additional small, poor-to-fair–quality 4-week prospective cohort study with high attrition (49%) reported an association between the pelvic examination and subsequent development of urinary symptoms (dysuria [11/63 vs 6/87; P < .01] and urinary frequency [17/63 vs 12/87; P < .01]).35 Further research is needed, in larger studies with urine culture–confirmed urinary tract infection as the outcome, to confirm or disprove this potential harm.
An overall summary of the evidence is presented in Table 6. There is no direct evidence examining the overall effectiveness of the pelvic examination in improving health outcomes for any of the gynecologic conditions included in this review. Cervical cancer, gonorrhea, and chlamydia were not included because other effective screening tests are already recommended for these conditions.5,6 Despite the many diseases that are plausibly detectable or that physicians cite as important to detect with routine screening pelvic examinations,36,37 this systematic review identified diagnostic accuracy studies for only 1 cancerous condition (ovarian cancer) and 3 infectious conditions (bacterial vaginosis, HSV, and trichomoniasis). The limited evidence identified on the accuracy of the screening pelvic examination for detecting any single gynecologic condition and the overall conclusion of its relatively low PPV for ovarian cancer are consistent with other systematic reviews17,18 and are not surprising, given that ovarian cancer is rare (approximately 12 cases per 100 000 women).38 Prior cross-sectional studies have shown that even under circumstances optimal for accurate disease detection (ie, patients examined preoperatively under anesthesia, all with some pelvic abnormality as indication for surgery, examination performed by attending physicians), the accuracy of the pelvic examination for detecting pelvic masses is low.39,40 Performance of the examination is notably worse when performed by inexperienced examiners and in patients who are obese or have enlarged uteri.39-42 To our knowledge, this review is the first systematic review to include accuracy data from the large PLCO trial confirming the very low sensitivity of ovarian palpation for the detection of ovarian cancer and estimating an 11% surgery rate resulting from abnormal examination findings.
The 4 studies reporting diagnostic accuracy for infectious diseases in high-risk settings (STI clinics or populations with high rates of symptoms) likely overestimate the accuracy obtained when pelvic examinations are administered to average-risk, asymptomatic primary care populations. Although the sensitivity of the pelvic examination was generally low for detecting a single condition (eg, ovarian cancer or bacterial vaginosis), on the basis of current evidence it is not possible to estimate the value of the screening pelvic examination to detect any of a host of conditions that might be found before symptomatic clinical presentation. In the absence of trials and test accuracy data on summary performance of screening pelvic examinations, clinical guidance could be informed by epidemiologic evidence to estimate the burden of single conditions, biological plausibility of early detection using the pelvic examination, treatability of the disease at earlier stages, and the presence of alternative, superior screening approaches.
The role of the annual or periodic screening pelvic examination is controversial.9,43-47 There are concerns that scrutiny of the accuracy of screening pelvic examination is misdirected, as the benefit of other routinely provided physical examination components (eg, heart and lung auscultation screening) in the context of screening is similarly lacking.44 Rather than viewing the examination as a screening test, this perspective suggests a broader clinical purpose, including to facilitate discussion of sensitive topics that would otherwise not be discussed.47 On the other hand, the screening pelvic examination can cause anxiety and discomfort and could pose unnecessary barriers to care,48 particularly in certain subpopulations of women (eg, those with a history of sexual abuse49-51).
Women’s preventive care is provided by clinicians with diverse training and professional orientations, and the type of clinician can influence the comprehensiveness of primary care.52-56 In the face of changing clinical practice with respect to cervical cancer screening frequency and new pelvic examination screening recommendations from professional organizations,14,57 long-standing patterns of women’s primary health care delivery may be altered. Research into the benefits or unintended consequences of different screening pelvic examination guidelines, ranging from recommendations against their provision14 to recommendations for shared decision making, is needed.15,57 Patients’ expectations and preferences for pelvic examination also warrant investigation, as it is not clear how shared decision making may change current clinical screening practices. Regardless of recommendations for screening pelvic examinations, it has been noted that women may wish to have an annual gynecological visit to ensure a full spectrum of preventive care.56
This review included English-language studies only, consistent with the USPSTF methods.16 This review excluded conditions such as gonorrhea, chlamydia, and cervical cancer, given that there are more accurate, alternative screening modalities for these conditions. Otherwise, despite having broad search terms to ensure an exhaustive literature search, the review identified little evidence to address the key questions.
No direct evidence was identified for overall benefits and harms of the pelvic examination as a 1-time or periodic screening test. Limited evidence was identified regarding the diagnostic accuracy and harms of routine screening pelvic examinations in asymptomatic primary care populations.
Corresponding Author: Janelle M. Guirguis-Blake, MD, Kaiser Permanente Research Affiliates EPC, Department of Family Medicine, University of Washington, 521 Martin Luther King Jr Way, Tacoma, WA 98405 (email@example.com).
Author Contributions: Dr Guirguis-Blake had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This research was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the USPSTF.
Role of the Funder/Sponsor: Investigators worked with USPSTF members and AHRQ staff to develop the scope, analytic framework, and key questions for this review. AHRQ had no role in study selection, quality assessment, or synthesis. AHRQ staff provided project oversight; reviewed the report to ensure that the analysis met methodological standards; and distributed the draft for peer review. Otherwise, AHRQ had no role in the conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript findings. The opinions expressed in this document are those of the authors and do not reflect the official position of AHRQ or the US Department of Health and Human Services.
Additional Contributions: We gratefully acknowledge the following individuals for their contributions to this project: Nadia Redmond, MSPH, Smyth Lai, MLS, and Elizabeth Hess, MA (Kaiser Permanente Center for Health Research); Barnett Kramer, MD, Paul Pinsky, PhD, and Maryam Doroudi, PhD, for providing the unpublished data from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial participants undergoing ovarian palpation; Tina Fan, MD (Agency for Healthcare Research and Quality); and current and former members of the USPSTF who contributed to topic deliberations. USPSTF members, peer reviewers, and federal partner reviewers did not receive financial compensation for their contributions.
Additional Information: A draft version of this evidence report underwent external peer review from 3 content experts (Hanna Bloomfield, MD, Minneapolis VA Health Care, University of Minnesota School of Medicine; David Chelmow, MD, VCU Medical Center; Giang Nguyen, MD, University of Pennsylvania) and 3 federal partners (Military Health System, National Institutes of Health, and Centers for Disease Control and Prevention). Comments were presented to the USPSTF during its deliberation of the evidence and were considered in preparing the final evidence review.
Editorial Disclaimer: This evidence report is presented as a document in support of the accompanying USPSTF Recommendation Statement. It did not undergo additional peer review after submission to JAMA.