Mark, Karen E. MD*†; Gunn, Robert A. MD, MPH†‡
GONORRHEA IS REPORTED passively in the United States by clinicians and, in most states, by laboratories to local public health departments, where data are reviewed, duplicates removed, and clean data forwarded through state health departments to the Centers for Disease Control and Prevention (CDC). Information collected on reported cases at the national level is limited to basic patient demographics: age, sex, and race/ethnicity, the latter of which is often missing. Information on diagnostic test used and behavioral risk factors such as sexual orientation is not collected, whereas information on anatomic site of infection and treatment is collected in some states, including California, but is also often missing.
With only basic demographic information consistently collected on reported cases, changes in gonorrhea rates over time can be difficult to interpret. San Diego County, California, experienced a 20% increase in reported gonorrhea cases from 1999 (1561 cases, 55 per 100,000 population) through 2001 (1875 cases, 64 per 100,000 population), with a widening male-to-female case ratio that could indicate increased transmission among men who have sex with men (MSM). Statewide data for California showed a 25% increase in cases over the same time period, 1 and national data showed the leveling off of a previous decline in cases. 2 We postulated that obtaining behavioral risk factor information on patients with gonorrhea might help to explain changes in gonorrhea rates over time, identify risk groups, and direct and monitor interventions. Furthermore, information on reason for testing and test type used to diagnose gonorrhea (culture, DNA probe test, or nucleic acid amplification test [NAAT]) would be useful over time to assess whether increases in gonorrhea rates might be the result of increases in screening or use of more sensitive tests (ie, NAATs) rather than true increases in disease. Finally, with the emergence of fluoroquinolone-resistant Neisseria gonorrhoeae (QRNG) in the continental United States, 3 identifying the proportion of gonorrhea diagnosed by culture (and thus amenable to susceptibility testing), and the antibiotics with which patients are being treated, is essential. Because this amount of information would be too burdensome to collect on every reported patient with gonorrhea, we developed a sample survey methodology that involves selecting a sample of patients with reported gonorrhea and collecting additional information from their healthcare providers. We then evaluated the feasibility and cost of using this method to enhance routine gonorrhea surveillance.
Sample patients were selected by including all gonorrhea patients reported to the San Diego County Health and Human Services Agency during the week of August 16, 2001, and then prospectively including all patients reported on alternate weeks until an estimated 110% of our desired sample size was reached (the week of October 18, 2001). No available information indicated that patients reported any one week were different from patients reported another week. Using Epi Info 6.04 (CDC, Atlanta, GA), we calculated that a sample size of at least 216 was needed to obtain 95% confidence intervals (CI) of ± 7% around a frequency of 50%, ± 4% around a frequency of 10%, and ± 3% around a frequency of 5%.
Healthcare providers caring for sample patients were contacted by telephone by communicable disease investigators and interviewed about sample patients’ demographics, risk factors, and management. Information on all anatomic sites tested for gonorrhea, and the results of each test, was collected, regardless of whether the test result was positive. Patients were not contacted, although likely could have provided more risk factor information; resources were unavailable to contact both clinicians and patients, and patients would be unlikely to provide accurate clinical information. When the diagnosing clinician was unavailable, information from the patient’s medical record was obtained from other office staff or through medical record review by disease investigators. At least 4 telephone attempts were made to contact each provider.
Disease investigators recorded the amount of time spent setting up and carrying out each interview or record review, including unsuccessful attempts. Data entry time was also recorded. To estimate total resources used, the average time (in hours) spent on each task was multiplied by health department salary and benefits cost per hour of staff time required. Supervision and data analysis time were not included. Data were entered in Epi Info 6.04 and analyzed in Epi Info 6.04 and Epi Info 2000. Prevalence ratios were used to compare characteristics of patients diagnosed in public sexually transmitted disease (STD) clinics with those diagnosed by private providers. 4 Private providers, who diagnose the majority of patients with gonorrhea but about whose practice least is known, were defined as all non-STD clinic providers except those practicing at government facilities (military, jail, or juvenile detention, where specific protocols are usually followed). This public health disease surveillance activity did not require Institutional Review Board approval. Standard public health patient confidentiality procedures were followed.
The initial sample consisted of 263 cases reported during 6 study weeks. Twelve (5%) cases did not have gonorrhea (most had chlamydia incorrectly recorded as gonorrhea), and 3 (1%) did not reside in San Diego County. Of the remaining 248 cases reported by clinicians and/or laboratories, information was obtained on 224 (90%). Reasons for lack of data included failure of the healthcare provider to return at least 4 telephone calls/refusal to participate (n = 19) and inability to locate the patient record (n = 5). Of the 224 patients on whom data were obtained, information was gathered by speaking with the clinician who diagnosed the patient with gonorrhea (30%), non-clinical office personnel (21%), nurses (18%), another physician, nurse practitioner, or a physician’s assistant (15%), or through medical record review by disease investigators (16%).
Demographic characteristics of the 224 sample patients with data obtained were similar to those of all reported San Diego gonorrhea patients in 2001 (Table 1). Fifty-five (25%) sample patients were diagnosed in STD clinics, 18 (8%) in other government facilities (military [n = 13], jail [n = 3], and juvenile detention [n = 2]), and 151 (67%) by private providers, specifically 41 (18%) in private offices, 37 (17%) in family planning clinics, 24 (11%) in emergency rooms or urgent care clinics, 16 (7%) in health maintenance organizations, 14 (6%) in community clinics, 11 (5%) in other private clinics, and 8 (4%) in inpatient or obstetric facilities. In San Diego, all these facilities are privately owned except for STD clinics and military, jail, and juvenile detention facilities.
Table 2 shows patient clinical characteristics and services received and compares patients diagnosed in public STD clinics with those diagnosed by private providers. STD clinic gonorrhea patients were similar to private provider gonorrhea patients in age and race/ethnicity (data not shown). Overall, gonorrhea patients were usually symptomatic (68%) and presented to their provider a median of 2 days (interquartile range, 0–4 days) after symptom onset. Twenty-six percent had a history of gonorrhea or chlamydia in the past 5 years. Patients were diagnosed by culture (40%), DNA probe (34%), and NAAT (21%), with 5% unknown. The majority of STD clinic patients were diagnosed by culture (75%), whereas private provider patients were more often diagnosed by DNA probe (45%). However, private provider patients accounted for almost half of all positive cultures. The detection of asymptomatic patients through screening with urine NAATs was uncommon (4% of all reported patients).
Cephalosporins were used to treat most (70%) private provider patients, whereas fluoroquinolones were used to treat most (89%) STD clinic patients. Among the private provider patients treated with cephalosporins, 39% (41 of 106) were treated with oral cefixime and the remainder with parenteral ceftriaxone; 49% (20 of 41) of the patients treated with cefixime received directly observed therapy. Twelve percent of private provider patients were treated with inappropriate antibiotics or not at all. STD clinic patients were more likely to have received directly observed therapy and other recommended preventive services, including empiric cotreatment for chlamydia, HIV testing, hepatitis B vaccination, and risk reduction counseling such as advice to decrease the number of sex partners and/or use condoms.
STD clinic and private provider partner management strategies were similar (data not shown). Overall, for 45% of patients, providers did not have any information about partner referral or treatment; for 38%, patient referral (having the patient tell their partner to get treated, or giving medication [n = 1] or a prescription [n = 3] directly to the patient to give to the partner) was done; for 12%, partner treatment was documented in the medical record; and for 5%, other strategies were used or the partner(s) were assumed already treated or known to be anonymous. Presumably because treatment was empiric and no follow up occurred after a positive test result, 21% of patients were not informed of their diagnosis, precluding them from notifying sex partners specifically about a gonorrhea exposure.
Among 143 men with gonorrhea, providers reported that 51 (36%) were MSM, 47 (33%) had sex only with women, and information on gender of sex partners was unknown for 45 (31%). Thus, at minimum, MSM constituted 23% (51 of 224) of all patients with reported gonorrhea. Sixty-seven percent (31 of 46) of male STD clinic gonorrhea patients and 24% (20 of 84) of male private provider patients were reported to be MSM (prevalence ratio [PR], 2.8; 95% CI, 1.8–4.4 but male private provider gonorrhea patients were more likely than male STD clinic gonorrhea patients to be of unknown sexual orientation (45% [38 of 84] vs. 4% [2 of 46]; PR, 10.4; 95% CI, 2.6–41.2). Rectal and pharyngeal cultures were rarely performed on MSM with gonorrhea reported by private providers, and no rectal or pharyngeal gonorrhea was diagnosed in the private sector (Table 3). However, in the STD clinic, where protocol recommends that all sexually exposed sites be tested, 23% of MSM with gonorrhea had rectal or pharyngeal infection in the absence of urethral infection.
Data collection was completed a mean of 35 days (range, 1–99 days) after gonorrhea report date and required a mean of 21 minutes (range, 5–60 minutes) of investigator time to complete (mean setup time, 7 minutes; interview, 14 minutes). As shown in Table 4, the estimated local resources required to implement this type of gonorrhea sample survey methodology were $2648 per year, or $12 per patient in the completed sample.
Our telephone survey of patients’ healthcare providers proved to be a feasible way to collect important additional information on a representative sample of all reported gonorrhea cases, and allowed us to estimate both characteristics of all reported cases and the burden of disease in specific high-risk populations. For example, applying the proportional distributions obtained from the sample, we estimated that there were at minimum 418 gonorrhea cases among MSM in San Diego County in 2001, assuming that men with unknown sexual orientation were not MSM. The use of sampling methodology to obtain population-based estimates of disease burden is not new. The National Health and Nutrition Examination Survey (NHANES) has been used to estimate the number and describe the characteristics of noninstitutionalized U.S. civilians with numerous health conditions, including herpes simplex virus type 2 5 and hepatitis C virus infection. 6 CDC’s hepatitis sentinel county study uses detailed epidemiologic data collected from persons with viral hepatitis in 5 representative counties to describe risk factors for viral hepatitis in the U.S. population. 7 These types of data are frequently cited and often used to estimate the prevalence of various risk factors and missed prevention opportunities among persons with a disease in order to guide prevention strategies and appropriately allocate prevention funds.
To our knowledge, this is the first published report to describe detailed characteristics of all patients with reported gonorrhea in a surveillance jurisdiction using a sampling methodology. Other investigators have described characteristics of predominantly asymptomatic gonorrhea patients detected through screening of individuals in probability-sampled households in Baltimore 8,9 and randomly selected low-income neighborhoods in northern California, 10 or have described characteristics of certain subgroups of gonorrhea patients such as adolescents 11 and men with urethral gonorrhea attending STD clinics. 12,13 However, the persons in these studies are not representative of all reported gonorrhea cases.
In our sample, gonococcal culture was used to diagnose 40% of gonorrhea patients: 75% of STD clinic and 25% of private provider patients. Although not directly comparable because of methodologic differences, a national sample of U.S. physicians surveyed in 2000 to 2001 (87% in private practice) reported a similar number of physicians (31%) “always” using culture to diagnose gonorrhea. 14 Maintaining local gonococcal culture capability is essential because the prevalence of antibiotic-resistant (particularly fluoroquinolone-resistant) gonorrhea continues to increase. 3 In our survey, only 4% of reported gonorrhea patients were asymptomatic and diagnosed by urine NAAT, indicating that urine NAAT-based screening is either still not frequently being done in the private sector or that, if done, few infections are found.
Recommended antibiotics for gonorrhea treatment have changed over time, because the organism has developed progressive antibiotic resistance. Our data, which show that 56% of patients were treated with cephalosporins and 36% with fluoroquinolones, were collected before the May 2002 recommendation by the California Department of Health Services and CDC that clinicians avoid using fluoroquinolones for gonorrhea treatment in California. 3,15 In 2002, the only pharmaceutical manufacturer of cefixime also decided to cease its production, 16 further complicating gonorrhea treatment. Thus, the more expensive parenteral ceftriaxone is the only remaining primary CDC-recommended gonorrhea treatment option in California. 16 Although protocol-driven STD clinics can quickly adapt to such changing recommendations, dissemination of information to private providers can take longer. 17 All clinicians who participated in the survey were mailed a letter describing the new recommendations in June 2002.
Nationwide trends in the proportion of male STD clinic gonococcal urethritis patients who are MSM have been described by using data from the Gonococcal Isolate Surveillance Project (GISP), which collects information on men with urethral gonorrhea diagnosed by culture in participating U.S. STD clinics. 13 In 2001, 50% of San Diego GISP patients were MSM, 12 whereas in our sample, 67% of male STD clinic gonorrhea patients were MSM. Our data indicate that GISP underestimates the proportion of STD clinic men with gonorrhea who are MSM by including only men with urethral gonorrhea and excluding the approximately one fourth of all gonorrhea-infected MSM who have rectal or pharyngeal gonorrhea in the absence of urethral gonorrhea. Drawing conclusions about the proportion of male gonorrhea patients in the private sector who are MSM is more difficult, because gender of sex partners was unknown for almost half of male private provider gonorrhea patients. However, at minimum, 24% of these patients were MSM. All patients being evaluated for STDs should be asked about gender of sex partners and sites of sexual exposure, including the oropharynx and rectum, and all exposed sites should be tested for gonorrhea.
CDC STD Treatment Guidelines 15 recommend that patients with gonorrhea be empirically treated for coinfection with Chlamydia trachomatis unless infection has been ruled out with a sensitive assay, that information about risk reduction be given (for example, decreasing the number of sex partners and/or using condoms), and that patients be offered HIV testing and hepatitis B vaccination. Data from the sample survey showed that, community-wide, although most patients received cotreatment for chlamydia, considerably more risk reduction information, HIV testing, and hepatitis B vaccination needs to be provided. STD clinic patients more frequently received each of these prevention services than did private provider patients.
Overall, no information was available regarding partner management for 45% of patients, whereas patient referral was done for 38%. Our patient referral estimate is likely an underestimate, because some clinicians might have told patients to tell partners to seek care but not recorded this instruction in the chart or recalled it later. The aforementioned nationwide physician survey found that 82% of physicians reported always instructing gonorrhea patients to tell partners to seek care for diagnosis and treatment, and 4% always gave gonorrhea patients medication to give to partners. 14 However, these estimates could be high because physicians tend to overstate their compliance with clinical guidelines. 18,19
One limitation of collecting data from healthcare providers is that data on sexual orientation are often incomplete. Thus, our finding that 36% of men with gonorrhea in San Diego have sex with men is likely an underestimate. Interviewing sample patients directly to obtain sexual orientation and other behavioral risk factor information would yield more complete data but would require greater resources because telephone interviews with clinicians or medical record review would still be required to obtain accurate clinical information. A comprehensive approach, using both provider and patient interviews, should be evaluated.
Our method of collecting information from healthcare providers on a sample of patients with reported gonorrhea could be expanded beyond San Diego County to several geographically diverse local or state health jurisdictions in a distribution representative of gonorrhea morbidity in the United States and could provide valuable nationwide estimates. A completed sample size of approximately 100 patients per jurisdiction, with at least 10 jurisdictions included and a random or systematic, representative sample of reported cases selected within each jurisdiction, would produce meaningful local and nationwide data. In a nationwide system, programs could be prewritten for sample patient selection and data analysis, limiting local resources required to do these tasks. Repeated every few years, this methodology would allow for trend monitoring, targeted interventions, and evaluation of these interventions. However implemented, a national gonorrhea sample survey surveillance system would improve our understanding of gonorrhea epidemiology, diagnosis, and treatment in the United States.
2. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2001. Atlanta: US Department of Health and Human Services, September 2002.
3. Centers for Disease Control and Prevention. Increases in fluoroquinolone-resistant Neisseria gonorrhoeae
—Hawaii and California, 2001. Morb Mortal Wkly Rep 2002; 51: 1041–1044.
4. Gregg MB, ed. Field Epidemiology. New York: Oxford University Press, 1996: 104.
5. Fleming DT, McQuillan GM, Johnson RE, et al. Herpes simplex virus type 2 in the United States, 1976 to 1994. N Engl J Med 1997; 337: 1105–1111.
6. Alter MJ, Kruszon-Moran D, Nainan OV, et al. The prevalence of hepatitis C virus infection in the United States, 1988 through 1994. N Engl J Med 1999; 341: 556–562.
7. Francis DP, Hadler SC, Prendergast TJ, et al. Occurrence of hepatitis A, B, and non-A/non-B in the United States. CDC sentinel county hepatitis study I. Am J Med 1984; 76: 69–74.
8. Turner CF, Rogers SM, Miller HG, et al. Untreated gonococcal and chlamydial infection in a probability sample of Baltimore adults. JAMA 2002; 287: 726–733.
9. Rogers SM, Miller HG, Miller WC, Zenilman JM, Turner CF. NAAT-identified and self-reported gonorrhea and chlamydial infections: Different at-risk population subgroups? Sex Transm Dis 2002; 29: 588–596.
10. Ruiz JD, Molitor F, McFarland W, et al. Prevalence of HIV infection, sexually transmitted diseases, and hepatitis and related risk behavior in young women living in low-income neighborhoods of northern California. West J Med 2000; 172: 368–373.
11. Ellen JM, Kohn RP, Bolan GA, Shiboski S, Krieger N. Socioeconomic differences in sexually transmitted disease rates among black and white adolescents, San Francisco, 1990 to 1992. Am J Public Health 1995; 85: 1546–1548.
12. Centers for Disease Control and Prevention. Sexually transmitted disease surveillance 2001 supplement: Gonococcal Isolate Surveillance Project (GISP) Annual Report—2001. Atlanta: US Department of Health and Human Services, 2002.
13. Fox KK, del Rio C, Holmes KK, et al. Gonorrhea in the HIV era: A reversal in trends among men who have sex with men. Am J Public Health 2001; 91: 959–964.
14. St. Lawrence JS, Montaño DE, Kasprzyk D, Phillips WR, Armstrong K, Leichliter JS. STD screening, testing, case reporting, and clinical and partner notification practices: A national survey of US physicians. Am J Public Health 2002; 92: 1784–1788.
15. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines 2002. Morb Mortal Wkly Rep 2002; 51( No. RR-6): 2, 8, 36.
16. Centers for Disease Control and Prevention. Discontinuation of cefixime tablets—United States. Morb Mortal Wkly Rep 2002; 51: 1052.
17. Gunn RA, Rolfs RT, Greenspan JR, Seidman RL, Wasserheit JN. The changing paradigm of sexually transmitted disease control in the era of managed health care. JAMA 1998; 279: 680–684.
18. Montaño DE, Phillips WR. Cancer screening by primary care physicians: A comparison of rates obtained from physician self-report, patient survey, and chart audit. Am J Public Health. 1995; 85: 795–800.
19. Weinberg MS, Gunn RA, Mast EE, Gresham L, Ginsberg M. Preventing transmission of hepatitis B virus from people with chronic infection. Am J Prev Med 2001; 20: 272–276.