Objective: To assess compliance with practice guidelines and to determine the extent of missed opportunities for sexually transmitted disease (STD) prevention by describing screening practices of a national sample of obstetricians and gynecologists and comparing them to the practices of other specialists.

Methods: Physicians (n = 7300) in five specialties that diagnose 85% of STDs in the United States were surveyed. Obstetrics and gynecology (n = 647) was one of the five specialties. Besides providing demographic and practice characteristics, respondents answered questions about who they screen (nonpregnant females, pregnant females) and for which bacterial STDs (syphilis, gonorrhea, chlamydia).

Results: Responding obstetricians and gynecologists were most likely to be non-Hispanic white (75%), male (66%), and in their 40s (mode 43 years old). They saw an average of 90 patients per week during 47 hours of direct patient care. Approximately 95% practiced in private settings. Almost all (96%) screened some patients for at least one STD. Obstetricians and gynecologists screened women more frequently than other specialties, but no specialty screened all women or all pregnant women.

Conclusion: Obstetricians and gynecologists screen women for STDs at a higher rate than other specialties represented in this study. Consistent with published guidelines, most obstetricians and gynecologists in our survey screened pregnant women for chlamydia, gonorrhea, and syphilis. Nonetheless, only about half of obstetricians and gynecologists screened nonpregnant women for gonorrhea or chlamydia, and fewer screen nonpregnant women for syphilis.

^aCenters for Disease Control and Prevention, Atlanta, Georgia, USA
^bBattelle Centers for Public Health Research and Evaluation, Seattle, Washington, USA
^cUniversity of Washington, Seattle, Washington, USA

This research was funded by the Centers for Disease Control and Prevention through contract no. 200-96-0599.

(Obstet Gynecol 2002:100:801-807. © 2002 by The American College of Obstetricians and Gynecologists.)

Bacterial sexually transmitted diseases (STDs), particularly chlamydia and gonorrhea, are the most commonly reported STDs in the United States.¹ These diseases frequently are asymptomatic in both men and women; thus, screening is justified to identify and treat individuals who might be infected but who do not report or recognize their risk for these highly prevalent, but curable, diseases. Reported syphilis has reached historic lows during the past decade²; nevertheless, congenital syphilis cases still occur and can be attributed to inadequate screening.^3,4 The high cost and inordinate burden of congenital syphilis cases justifies screening for this low-prevalence disease. Intensive syphilis screening is also justified by the current effort to eliminate syphilis from the United States⁵ because increased screening will be necessary to identify and treat the remaining reservoir of undetected cases. In addition, estimates of national base rates by screening each of these diseases are needed to inform surveillance estimates.^5,6 This paper describes the STD screening practices of United States obstetricians and gynecologists who participated in a national survey.

Reasons to assess the screening practices of obstetricians and gynecologists include, first, the fact that sequelae of many STDs are worse for women than for men. For many women, obstetricians and gynecologists are their primary provider. Every encounter with an obstetrician or gynecologist provides a natural opportunity for STD screening within the context of a prenatal or annual reproductive health care visit. Second, women often do not know they are infected, perhaps because of the asymptomatic presentation of some STDs or failure to recognize symptoms. These untreated STDs can culminate in pelvic inflammatory disease and infertility. Finally, the consequences of untreated STD infections in pregnant women can be dangerous to the fetus and newborn infant. Thus, screening of pregnant women by obstetricians and gynecologists takes on added importance.⁷

The importance of screening for syphilis, gonorrhea, and chlamydia arises repeatedly in treatment guidelines issued by federal agencies and professional organizations, although most of these recommendations are much stronger with regard to screening of pregnant women. The United States Preventive Services Task Force and 2002 Centers for Disease Control and Prevention (CDC) treatment guidelines recommend screening all sexually active women to age 25 for chlamydia (and older women if risk factors are present), but offer less guidance about chlamydia screening for all women or for men.⁸ Both the United States Preventive Services Task Force and the American College of Obstetricians and Gynecologists recommend screening all pregnant women for common STDs such as chlamydia.^8,9 CDC treatment guidelines recommend universal screening for syphilis, but base screening recommendations for chlamydia and gonorrhea in pregnant women on considerations such as age, history of risk behavior, and prevalence of disease in the geographic area. The peer-reviewed scientific literature is consistent with this recommendation and promotes screening of pregnant women, especially for syphilis.^10,11

Most of the screening literature is based on small and local, rather than national, data. Warner et al¹² found a prevalence of 8.2 congenital syphilis cases per 1000 live births in Georgia after reviewing 157 charts. Schulte et al¹³ reported the prevalence of syphilis among women infected with human immunodeficiency virus in Texas and the number of congenital syphilis cases (51%) that followed births to these same women. When Mills et al¹⁴ asked 96 Minnesota obstetricians and family physicians for their screening endorsements and practices, virtually all of them (97%) endorsed prenatal screening for syphilis but only one in four endorsed prenatal screening for chlamydia (26%) or gonorrhea (24%). Recurring themes in this research literature include a focus on syphilis rather than other STDs, greater emphasis on screening of pregnant women, and statements about missed opportunities to detect STD. The limitations inherent in these reports of suboptimal screening, assessed primarily for a single STD and only in local areas support the need to conduct a nationally representative survey.

During 1999 and 2000, the CDC, in conjunction with the Battelle Centers for Public Health Research and Evaluation, conducted a national survey of physicians.¹⁵ Included in this survey were questions about screening practices for syphilis, gonorrhea, and chlamydia. We investigated the following questions: 1) What are obstetricians' and gynecologists' current screening practices for syphilis, gonorrhea and chlamydia? 2) Do these practices differ for pregnant versus nonpregnant women? 3) How do obstetricians and gynecologists differ from other physicians who see STDs in their practices with respect to screening of nonpregnant women?

Materials and Methods

Five medical specialties (obstetrics and gynecology, internal medicine, family practice or general practice, emergency medicine, and pediatrics) were selected for participation based on evidence that these specialties provide care for 85% of STDs diagnosed in the United States.^16,17 The pool of eligible physicians was identified using the American Medical Association's Physician Master File (a comprehensive list of physicians in the United States and its territories). Therefore, every doctor fulfilling our inclusion criteria had an equal chance of being selected: the relative proportions of physician specialties in the survey matched those in the American Medical Association's Master File. Further inclusion criteria that elevated the odds that participating physicians would see STDs were that they spent 50% or more of their time in direct patient care and provided care for patients between the ages of 13 and 60 years. Institutional review boards at the CDC, Battelle, and the United States Office of Management and Budget approved the survey.

Surveys were mailed by Federal Express to 7300 physicians selected at random from the five chosen specialties. Each survey included a $15.00 cash incentive and a postage-paid return envelope. Mailing parameters and incentive level were chosen on the basis of a previous test of the relative effects of mailing method and incentive size.¹⁸ A reminder card was mailed 10 days after the initial mailing, and repeat surveys were mailed to nonrespondents 4, 7, and 15 weeks after the initial mailing. After adjustments for surveys marked as undeliverable or returned as ineligible for reasons such as the physicians' retirement, the 4223 respondents corresponded to a 70.2% return rate. There were minor variations in the return rates from different specialties (64% for internists to 78% for emergency medicine). Although return rates were statistically different, ² (4) = 19.47, P < .05, the magnitude of the association, .07, is small. Geographically, respondents came from all 50 states and the District of Columbia, with regional distribution from the Northeast (21%), South (32%), Midwest (25%), and West (22%). Of the 4223 respondents, 90 reported they did not see STDs in their practices and were eliminated, leaving 4133 physicians for analyses. Six hundred fifty-six respondents reported obstetrics and gynecology as their primary specialty: nine were ineligible because they spent insufficient time seeing patients, leaving 647 in the final sample of obstetricians and gynecologists.

Descriptive analyses identified the characteristics of respondents; frequencies were used for categoric (eg, practice location) variables and means for continuous variables. Obstetricians and gynecologists were compared with physicians in the remaining four specialties by using ² tests and f coefficients to estimate effect sizes for differences. The f coefficient¹⁹ approximates the correlation coefficient for nominal data, such as the differences in screening patterns. Differences in screening rates for pregnant versus nonpregnant women by obstetrician-gynecologists were assessed using the McNemar ² test for related samples, a nonparametric equivalent to repeated measures tests for continuous variables. The related samples in this study would be the responses of physicians for screening different categories of women or different STDs.

Results

Table 1 presents descriptive data for the 647 obstetricians and gynecologists who responded to the survey and who fit the eligibility criteria. About two-thirds of physicians were male, and the modal practitioner was 43 years old with 15 years of professional experience. Most were in solo practices (33%) or in single-specialty settings (50%). Eighty percent of2 the obstetricians and gynecologists who responded to the survey practiced in primary care office settings, with most of the remainder practicing in hospital outpatient clinics. Virtually all physicians (94%) reported being in private practice rather than in a publicly funded practice setting. As expected, their patients were almost exclusively female (98%). Physicians' estimates of patient ages indicated slight positive skewness (skewness = 0.1, normal = 0, critical value = 0.6) and mild platykurtosis (kurtosis = 2.1, normal = 3.0, critical value = 1.8), but indicators of both deviations from normality were within acceptable boundaries for statistical analyses.

Obstetricians and gynecologists were more likely to screen pregnant women than nonpregnant women for STDs. Approximately four-fifths of them screened pregnant women for chlamydia (79.4%) and gonorrhea (79.6%), and 85.6% screened pregnant women for syphilis. About one in two screened nonpregnant women for chlamydia and gonorrhea and one in five for syphilis (Table 2). Each of these differences in proportions (pregnant versus nonpregnant patients, assessed by McNemar ²) was statistically significant. For chlamydia, ² (1) = 99.79; for gonorrhea, ² (1) = 129.52; for syphilis, ² (1) = 381.45, all P < .001. Thus, it is clear that obstetrician-gynecologists are far more likely to screen pregnant women than nonpregnant women for STDs.

We also tested whether the number of patients seen per week (an index of patient load) was correlated with screening for any of the three STDs, because it is plausible that a high patient load might preclude adequate time for optional measures such as screening. The resulting point-biserial correlations were negative and nonsignificant, indicating there was no relationship between screening frequency and patient load. Nor were physicians who see a higher proportion of young patients more likely to screen, despite the higher prevalence of chlamydia and gonorrhea in adolescents and young adults. Neither the proportion of 13- to 25-year-old patients in a practice nor the absolute number of patients seen in a week correlated significantly with screening for chlamydia, gonorrhea, or syphilis (all P > .05). Moreover, obstetricians' and gynecologists' screening practices were not associated with community size as defined in Table 1, (² [6] = 9.76, P > .10), although those in some states were more likely to screen than those in others (² [49] = 73.64, P < .05). Finally, female obstetricians and gynecologists (98.6%) screened for any STD at slightly greater rates than male obstetricians and gynecologists (95.0%) (² [1] = 4.83, P < .05), although the size of the association, .09, was small.

Obstetricians and gynecologists screened pregnant women for syphilis at nearly the same rates as for chlamydia and gonorrhea. Although a pairwise comparison using the McNemar ² statistic for the chlamydia screening rate (79.4%) versus the syphilis screening rate (85.6%) was significant (² [1] = 16.18, P < .001), the absolute rates are close to one another. The differences are much greater and in the opposite direction when nonpregnant women are concerned, however, for screening of chlamydia (54.6%) versus syphilis (22.9%) (² [1] = 199.04, P < .001). Gonorrhea and chlamydia screening rates for nonpregnant women also were similar to one another, and the comparison of gonorrhea with syphilis screening was comparable to the comparison of chlamydia with syphilis.

Virtually all (over 99%) of the physicians, regardless of specialty, treated women in their practices. Table 2 compares obstetricians' and gynecologists' reports of screening nonpregnant women against those reported by all other specialists aggregated, as well as the other four specialties individually. Across all three STDs (syphilis, gonorrhea, and chlamydia), obstetrician-gynecologists were more likely to screen nonpregnant women (23% to 55%, depending on disease) than were other specialists (19% to 31%). Notably, the lowest screening rate for the other specialties was for syphilis (19%). Thus, screening of nonpregnant women appears to be uncommon, despite the frequently asymptomatic nature of these diseases in women and the high long-term costs of undetected infections.

Considering the other specialties individually, Table 2 shows that family and general practitioners came closest to obstetricians and gynecologists in their screening practices, with rates of 21% to 39% for the STDs individually, and with two-thirds of family and general practitioners screening for any STD. Emergency medicine physicians were least likely to screen for syphilis, whereas internists were least likely to screen for gonorrhea and chlamydia.

Discussion

Although screening of pregnant women was more common than screening of nonpregnant women, screening of pregnant women was still well below the universal screening recommendations in the professional guidelines. The percentages in Table 2 reflect the percentage of physicians who reported screening; the proportion of patients who were actually screened might be lower than the percentages reported in Table 2. (For example, we report that 79.8% of obstetricians and gynecologists screened pregnant women for chlamydia. If this 79.8% screened two-thirds of their pregnant patients, then only about half of the pregnant patients [66.7% of 79.8% = 53.2%] would have been screened.) Thus, the figures in Table 2 provide an upper bound estimate for the proportion of patients who were screened (although some nonpregnant women could have been screened during an earlier pregnancy).

The physicians in this nationally representative sample were less likely to screen nonpregnant women than pregnant women for any STD. Obstetricians and gynecologists certainly screen nonpregnant patients with much more fidelity than do other specialties. However, only 22% to 54% of obstetricians, depending on disease, screened nonpregnant women. Physicians of other specialties who participated in our survey were even less likely to screen for bacterial STDs. There might be defensible reasons why physicians should not be encouraged to screen all women for a broad range of STDs. For example, physicians with older and monogamous clients in a low-prevalence location might believe that the benefits of screening are not worth the costs (and published guidelines support that position). Thus, screening rates for nonpregnant women that include many older women who are less at risk for STD acquisition are legitimately lower than screening rates for pregnant women.

Conversely, because younger women have higher prevalences of chlamydia and gonorrhea, if this were the case, we would expect to see higher rates of screening for these diseases by physicians with younger patients. In this survey, there was no such correlation, although, given the large sample size, we would have detected even a very small effect (eg, r < .10). Thus, it does not appear that this type of heuristic is guiding decisions whether to screen.

All professional guidelines are consistent in recommending universal screening of pregnant women. Most obstetricians and gynecologists who participated in this survey did, in fact, screen pregnant women. The findings from this study provide an interesting contrast to the results of a Georgia survey that reported screening rates of 71% (gonorrhea and chlamydia) to 98% (syphilis) for pregnant women.²⁰ In this nationwide sample, a lower percentage of obstetricians and gynecologists (85%) reported screening pregnant women for syphilis. Despite the relatively high proportion of obstetricians and gynecologists who screened pregnant women for STDs, there is still room for improvement in achieving the recommendations for universal screening of pregnant women.

Clearly, the less frequent screening for syphilis might make clinical sense because syphilis is much less common than the other two STDs. Currently, there are roughly 35 gonorrhea cases and 65 chlamydia cases reported for every case of syphilis.¹ However, given the inordinately high risk to the fetus or neonate of untreated syphilis, even in a low prevalence setting and in the midst of a national campaign to eliminate syphilis,⁵ this is one disease for which universal screening of pregnant women might be justified. Low screening rates of pregnant women by other specialties might not be problematic as long as pregnant women routinely are referred to obstetricians who are more likely to screen appropriately.

In comparing obstetricians and gynecologists with the other physicians, some of the other physicians practice in settings where STD screening is not the norm. Survey results cohere with likely practice expectations, as family and general practitioners were more likely to screen than internists, emergency medicine physicians, and pediatricians. Because guidelines typically suggest screening as a part of routine health care and most people are more likely to seek routine care from family and general practitioners than from emergency medicine physicians, the higher screening rates among family and general practitioners is logical. Should the differences in screening rates between obstetricians and gynecologists and the remaining specialties other than family and general practitioners be considered problematic? On the one hand, current practice standards do not encourage routine screening by some specialties, eg, emergency medicine physicians. On the other hand, screening in emergency rooms has identified a substantial number of previously undetected cases.^21,22 Thus, current practice standards that do not encourage STD screening miss many screening opportunities to detect sexually transmitted diseases. With roughly a third of emergency medicine physicians screening for at least one STD, however, we suggest there is a reasonable platform on which to build a systematic approach to screening emergency room patients. As for pediatricians and internists, the fact that they are among the specialties that see 85% of STDs^16,17 suggests that, although they may have many patients at very low risk (eg, prepubescent children), they do see some patients at higher risk. For example, many adolescents see pediatricians, and adolescents have some of the highest STD rates in the country.¹ Reinforcing this point is the fact that four-fifths of the respondents in this survey reported that they worked in a primary care office setting, ie, one in which a patient might expect to seek routine health care (and screening).

There are other control and prevention strategies beyond STD screening of pregnant and nonpregnant women that can be considered to reduce the STD burden in the United States. For example, screening of men could also disrupt transmission patterns. However, physicians infrequently screen men for STDs. Physicians in our sample who saw men in their practices had very low rates of STD screening (chlamydia 13%, gonorrhea 14%, syphilis 19%). Thus, men remain a potential reservoir of STDs for women, a point that has been made in print at least as far back as 1979.²³ As the United States Preventive Services Task Force notes, there is still insufficient information on the efficacy of male screening to guide formal screening recommendations. Clearly, the potential benefits and cost effectiveness of male screening needs further research.

There are several limitations to this research. One limitation is that only physicians' reported behaviors were described. The data do not identify the actual proportion of patients in a practice being screened, only that the physician indicated that screening occurred. We also addressed only curable bacterial STDs, not viral STDs, which might have had a different pattern of results. Finally, the research relied on physicians' self-reports of their practice characteristics and clinical behavior. These limitations suggest several avenues for further research. Focusing on patient-level data, including common viral STDs, and assessing the screening practices of other disciplines that provide obstetric care would be worthwhile.

One final point is that female obstetricians and gynecologists screened patients at slightly higher rates than male obstetricians and gynecologists. Because the proportion of women in medical practice (this is all physicians, not obstetricians and gynecologists specifically) has increased from 15.8% of all physicians in 1983 to 27.9% in 1999,²⁴ screening by obstetricians and gynecologists might increase further in the future because of changes in the specialty's gender distributions (assuming screening patterns among female physicians do not change). Nevertheless, these results underscore that there are many missed opportunities for STD screening and that the number of physicians who are screening both pregnant and nonpregnant women for STDs is below optimal levels regardless of their practice specialty.

References

1. Centers for Disease Control and Prevention. Tracking the hidden epidemics: Trends in STDs in the United States 2000. Washington, DC: U.S. Government Printing Office, 2001.

2. Centers for Disease Control and Prevention. Primary and secondary syphilis - US 1998. MMWR Morb Mortal Wkly Rep 1999;48:873-8.

3. Centers for Disease Control and Prevention. Congenital syphilis - US 1998. MMWR Morb Mortal Wkly Rep 1999;48:757-61.

4. Southwick KL, Guidry HM, Weldon MM, Mertz KJ, Berman SM, Levine WC. An epidemic of congenital syphilis in Jefferson County, Texas, 1994-1995: Inadequate prenatal syphilis testing after an outbreak in adults. Am J Public Health 1999;89:557-60.

5. Gayle HD, Counts GW. Syphilis elimination: A unique time in history. J Am Med Womens Assoc 2001;56:2-3.

6. Centers for Disease Control and Prevention. The national plan to eliminate syphilis from the United States. Washington, DC: U.S. Government Printing Office, 1999.

7. Holmes KK, Sparling PF, Mardh P, Lemon SM, Stamm WE, Piot P, et al. Sexually transmitted diseases. 3rd ed. New York: McGraw-Hill, 1999.

8. U.S. Preventive Services Task Force. Screening for chlamydial infection. Am J Prev Med 2001;20:90-4.

9. American College of Obstetricians and Gynecologists. Guidelines for perinatal care. 4th ed. Washington, DC: American College of Obstetricians and Gynecologists, 1997.

10. Nelson HD, Helfand M. Screening for chlamydial infection. Am J Prev Med 2001;20:95-107.

11. Genc M, Ledger WJ. Syphilis in pregnancy. Sex Transm Infect 2000;76:73-9.

12. Warner L, Rochat RW, Fichtner RR, Stoll BJ, Nathan L, Toomey KE. Missed opportunities for congenital syphilis prevention in an urban southeastern hospital. Sex Transm Dis 2001;28:92-8.

13. Schulte JM, Burkham S, Hamaker D, St. Louis ME, Paffel JM, Hutcheson D, et al. Syphilis among HIV-infected mothers and their infants in Texas from 1988 to 1994. Sex Transm Dis 2001;28:315-20.

14. Mills WA, Martin DL, Bertrand JR, Belongia EA. Physicians' practices and opinions regarding prenatal screening for human immunodeficiency virus and other sexually transmitted diseases. Sex Transm Dis 1998;25:169-75.

15. St Lawrence JS, Montano D, Kasprzyk D, Phillips W, Armstrong K, Leichliter J. National survey of US physicians' STD screening, testing, case reporting, clinical management, and partner notification practices. Am J Public Health. In press.

16. Hammett TM, Kaufman JA, Faulkner AH, Hoaglin DC, Battaglia MP, Felsenstein D, et al. Sexually transmitted disease prevention in the United States: Integrated evaluation of public and private sector disease reporting and service delivery. CDC 200-93-0633, Phase I Final Report. Atlanta: Centers for Disease Control and Prevention, 1997.

17. Department of Health and Human Services. Department of Health and Human Services; National Center for Health Statistics, data from the National Health Care Survey, 1995: Two parts: 1) National Ambulatory Medical Care Survey, 2) National Hospital Ambulatory Medical Care Survey. Washington, DC: U.S. Government Printing Office, 1997.

18. Kasprzyk D, Montano DE, St. Lawrence JS, Phillips WR. The effect of variations in mode of delivery and monetary incentive on physicians' responses to a mailed survey assessing STD practice patterns. Eval Health Professions 2001;24:3-17.

19. Hays WL. Statistics. 4th ed. Forth Worth, TX: Holt, Reinhart, and Winston, 1988.

20. Weisbord JS, Koumans EH, Toomey KE, Grayson C, Markowitz LE. Sexually transmitted diseases during pregnancy: Screening, diagnostic, and treatment practices among prenatal care providers in Georgia. South Med J 2001;94:47-53.

21. Mehta SD, Rothman RE, Kelen GD, Quinn TC, Zenilman JM. Clinical aspects of diagnosis of gonorrhea and chlamydia infection in an acute care setting. Clin Infect Dis 2001;32:655-9.

22. Mehta SD, Shahan J, Zenilman JM. Ambulatory STD management in an inner-city emergency department: Descriptive epidemiology, care utilization patterns, and patient perceptions of local public STD clinics. Sex Transm Dis 2000;17:154-8.

23. Felton WF. A theory of the epidemiology of gonorrhea. Br J Venereal Dis 1979;55:58-61.

24. Department of Commerce. Statistical abstract of the United States, 2001. Washington, DC: U.S. Government Printing Office, 2001.

Address reprint requests to: Matthew Hogben, PhD, Division of STD Prevention, Mail Stop E-44, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333; E-mail: mhogben@cdc.gov

Received February 25, 2002.
Received in revised form May 8, 2002.
Accepted May 16, 2002.