Compliance with the Centers for Disease ... - Wiley Online Library

7 downloads 14889 Views 79KB Size Report
comply with the Centers for Disease Control and Prevention. (CDC) recommendations for .... criteria category are denoted in Table 1 by an asterisk. For example, under ..... However, the guidelines call for the discretionary addition of Flagyl to ...
ACAD EMERG MED

d

April 2004, Vol. 11, No. 4

d

371

www.aemj.org

Compliance with the Centers for Disease Control and Prevention Recommendations for the Diagnosis and Treatment of Sexually Transmitted Diseases Bryan G. Kane, MD, Linda C. Degutis, DrPH, Helen K. Sayward, MS, Gail D’Onofrio, MD, MS Abstract Little is known about gaps in quality and the extent to which clinical standards are used in emergency department (ED) practice. Objectives: To determine whether ED practitioners comply with the Centers for Disease Control and Prevention (CDC) recommendations for diagnosing and treating sexually transmitted diseases (STDs). Methods: A retrospective chart review of ED visits was conducted at an urban teaching hospital. Using ICD-9 codes, urethritis, cervicitis, pelvic inflammatory disease (PID), gonorrhea, and chlamydia cases seen from May 1, 2000, to February 28, 2001, were identified. Documentation of components of the history, physical examination, diagnostic testing, prescribed antibiotics, and discharge instructions necessary to comply with the CDC guidelines were abstracted. This set of comprehensive criteria was compared with a less stringent subset of selected criteria. Results: Two hundred forty-six patient visits were identified, and 203 (83%) were included. Forty-eight men and 155 women were included: 48 (24%)

with urethritis, 34 (17%) with cervicitis, and 121 (60%) with PID. For urethritis, cervicitis, and PID, respectively, there was documentation of compliance with indicators related to the following: history 73%, 15%, and 14%; physical examinations 63%, 15%, and 22%; diagnostic testing 79%, 71%, and 71%; antibiotic use 33%, 32%, and 32%; and safe sex instructions 50%, 18%, and 15% of the time. Men were more likely to receive safe sex instructions (p # 0.01). Total (100%) compliance in all five domains occurred 8% of the time for urethritis, 3% for cervicitis, and never for PID. The rates of 100% compliance were not significantly different when a subset of selected criteria was used. Conclusions: Deficits in adherence to recommended guidelines for the diagnosis and treatment of STDs exist in ED practice. Key words: quality measures; clinical guidelines; sexually transmitted diseases (STDs); urethritis; cervicitis; pelvic inflammatory disease. ACADEMIC EMERGENCY MEDICINE 2004; 11:371–377.

Recent evidence of suboptimal quality in clinical medicine, posing serious threats to the health of the American public, has highlighted the need for strategies to reduce these deficits.1 Many adjuncts available to providers, such as ‘‘computer-based decision support,’’2 computer-generated reminders of guidelines,3 databases,4 and targeting evidence at the point of care,5 show some promise. In 1993, Grimshaw and Russell6 reviewed the existing literature on clinical guidelines and concluded that their use leads to a reduction in medical error.

However, the mere presence of clinical guidelines does not ensure use in daily clinical practice. For example, a retrospective chart review of Advanced Cardiac Life Support (ACLS) interventions found a 64.8% compliance rate.7 Many barriers to implementation exist relating to both physician and system issues. Hardern and Hampshaw8 surveyed emergency medicine practitioners and found that despite positive attitudes toward clinical guidelines, factors such as being ‘‘user-friendly’’ mitigate the impact they might have. In another study, work environment and workload were identified by physicians as causes of prescription error.9 These challenges are faced daily by emergency medicine practitioners. Despite efforts by such groups as the American College of Emergency Physicians (ACEP), which generates practice guidelines for specific aspects of emergency medical care, the existence of such has not been shown to alter practitioner documentation or actions.10 Currently, the majority of the literature assessing health care practitioner compliance with guidelines is measured through survey responses. However, the perceived importance and use of guidelines by practitioners may not reflect actual performance. This overestimation of use and adherence with clinical

From the Department of Surgery, Section of Emergency Medicine, Yale University School of Medicine, New Haven, CT (BGK, LCD, HKS, GDO). Received June 13, 2003; revision received October 28, 2003; accepted November 14, 2003. Presented at the SAEM New England Regional Meeting, Worcester, MA, April 2002, and the Connecticut College of Emergency Medicine Annual Meeting, November 2001. Address for correspondence: Gail D’Onofrio, MD, MS, Section of Emergency Medicine, 464 Congress Avenue, Suite 260, New Haven, CT 06519. Fax: 203-785-4580; e-mail: [email protected]. Address for reprints: Bryan G. Kane, MD, Lehigh Valley Hospital, Department of Emergency Medicine, Cedar Crest and I-78, P.O. Box 689, Allentown, PA 18105-1556. Fax: 610-402-7160; e-mail: bryan. [email protected]. doi: 10.1197/j.aem.2003.11.016

372

Kane et al.

guidelines by survey methods was previously reported.11 For example, when actual performance was examined, such as in an investigation into the use of antibiotics for urinary tract infections, it was found that prescriptions for the more costly, although not recommended, fluoroquinolones was increasing.12 More recently, McGlynn et al.1 evaluated performance on 439 indicators of quality of care for 30 acute and chronic conditions as well as preventive care. Their results indicate that Americans receive only about half of recommended medical care processes.1 We chose to measure practitioner compliance with a well-known, nationally accepted set of clinical guidelines, the 1998 Centers for Disease Control and Prevention (CDC) ‘‘Guidelines for the Diagnosis and Treatment of Sexually Transmitted Diseases.’’13 These guidelines are clear in their diagnostic criteria, obtained through patient history, physical examination, and laboratory testing. They also contain treatment protocols, disposition criteria, and discharge instructions. The objective of this study was to quantify health care practitioner compliance with CDC-recommended diagnostic and treatment criteria for urethritis, cervicitis, and pelvic inflammatory disease (PID) in the emergency department (ED) setting.

detailed in the CDC’s ‘‘1998 Guidelines for Treatment of Sexually Transmitted Diseases.’’13 The specific components recommended by the CDC for each of five domains, including 1) history, 2) physical examination, 3) diagnostic testing, 4) antibiotic treatment, and 5) discharge instructions, are provided in Table 1. Only diagnostic criteria clearly used and available in an ED setting were deemed essential for providers to document and were included. For example, demonstration of endometritis on endometrial biopsy is a ‘‘definitive criterion’’ to diagnose PID but is not part of the scope of the practice of emergency medicine. We used all components in each domain recommended by the CDC and outlined in Table 1 as our comprehensive (full) criteria. We chose a subset of the comprehensive criteria as essential to the domains of history and physical examination and created a selected criteria category. Those items included in the selected criteria category are denoted in Table 1 by an asterisk. For example, under cervicitis, the CDC recommends five items to be part of the physical examination. However, under the selected criteria we included only three items. The domains entitled diagnostic testing and discharge instructions remained the same. With respect to antibiotic use, if the correct antibiotic (i.e., ceftriaxone) was given at a higher dose than recommended for urethritis and cervicitis (250 mg vs. 125 mg), the practitioner was given credit for compliance in the selected criteria. Demographic information, including age, gender, and chief complaint, was recorded for all visits. All pertinent points of the history and physical examination were recorded as present, absent, or not documented. The same format was used for urine human chorionic gonadotropin (hCG) testing. Cultures for gonorrhea and chlamydia were recorded as sent or not sent. All patients had the empiric treatment they received recorded. Disposition was recorded as admission or discharge, and if discharged, instructions were recorded as given or not given.

METHODS Study Design. This retrospective chart review was conducted from March 1, 2000, through February 28, 2001. The study was approved by the institution’s human investigation committee. Study Setting and Population. This study was performed at an urban, tertiary care, teaching hospital, with a census of approximately 60,000 visits per year. Patients 14 years of age or older who were seen in the ED with a discharge diagnosis of gonorrhea (Gc), chlamydia (Ch), cervicitis, urethritis, or PID were included. Cases were initially identified by using International Classification of Diseases (ICD-9) discharge diagnostic codes. The specific ICD-9 codes used to identify such encounters were 089.00 through 089.89 for gonorrhea; 099.40 and 099.53 through 099.55 for chlamydia; 597.80, 597.81, and 597.89 for urethritis; 614.90 for PID; and 616.0 for cervicitis. Individual patients with multiple ED visits in the enrollment period were eligible to be enrolled once for each disease episode. Cases were excluded after reviewing the chart if the coding was determined to be incorrect or if documentation of the ED visit identified by ICD-9 codes was missing from the medical record. Study Protocol. Data were extracted from the medical record according to components specifically

d

COMPLIANCE WITH CDC FOR STD TREATMENT

Data Analysis. All cases were analyzed according to the provider’s diagnosis. For example, if a provider’s diagnosis was PID, the patient’s chart was held to the CDC’s standards for that disease. When conflict was noted between resident or physician assistant records and the attending physician notes, the attending physician record was used. If conflict between the practitioner’s pharmacologic orders and nursing documentation of the drug administration was noted, documentation of the drug delivered from the nursing note was used. If data points were not included in the provider notes, other written records of the encounter contained in the permanent medical record were used. This included use of nursing notes to determine patient temperature, hCG testing, documentation of whether cultures were sent, and printed computer orders and laboratory results.

ACAD EMERG MED

d

April 2004, Vol. 11, No. 4

d

373

www.aemj.org

TABLE 1. Centers for Disease Control and Prevention (CDC) Recommendations: Essential Components for Each Domain Urethritis History

Presence or absence of: 1. Urethral discharge* 46/48 (96%) 2. Dysuria 36/48 (75%)

Cervicitis Presence or absence of: 1. Vaginal discharge* 25/34 (74%) 2. Abnormal vaginal bleed 10/34 (29%) 3. Lower abdominal pain* 23/34 (68%)

PID Presence or absence of: 1. Lower abdominal pain* 112/121 (93%) 2. Vaginal discharge* 75/121 (62%) 3. Abnormal vaginal bleeding 44/121 (36%) 4. Dyspareunia 13/121 (11%)

Physical examination

1. Mucopurulent or purulent discharge* 30/48 (63%)

1. Endocervical discharge or friable cervix* 25/34 (74%) 2. Temperature \1018F 31/34 (91%) 3. No lower abdominal tenderness 11/34 (32%) 4. No adnexal tenderness* 29/34 (85%) 5. No cervical motion tenderness* 12/34 (35%)

1. Lower abdominal tenderness* 105/121 (87%) 2. Adnexal tenderness* 39/121 (32%) 3. Cervical motion tenderness* 89/121 (74%) 4. Temperature [1018F 116/121 (96%) 5. Endocervical discharge 93/121(77%)

Diagnostics

1. GC/Ch culture sent* 38/48 (79%)

1. GC/Ch cultures* 26/34 (77%) 2. Urine b-hCG* 29/34 (85%)

1. GC/Ch cultures* 93/121 (77%) 2. Urine b-hCG* 105/121 (87%)

Antibiotics

Discharge instructions

Empiric treatment for both GC and Ch is necessary 16/48 (33%) 1. GC: cefixime 400 mg PO, or ceftriaxone 125 mg IM, or ciprofloxacin 500 mg PO, or ofloxacin 400 mg PO 2. Ch: Azithromycin 1 g PO, or doxycycline 100 mg PO bid for 7 days, or ofloxacin 300 mg PO bid for 7 days or multiple erythromycin regimens 1. Safe sex instructions* 24/48 (50%)

Empiric treatment for both GC and Ch is necessary 11/34 (32%) 1. GC: cefixime 400 mg PO, or ceftriaxone 125 mg IM, or ciprofloxacin 500 mg PO, or ofloxacin 400 mg PO 2. Ch: Azithromycin 1 g PO, or doxycycline 100 mg PO bid for 7 days, or ofloxacin 300 mg PO bid for 7 days or multiple erythromycin regimens

1. Safe sex instructions* 6/34 (18%)

All outpatients must be treated by Regimen A or Regimen B 28/88 (32%) 1. Regimen A: ofloxacin 400 mg PO bid for 14 days, and metronidazole 500 mg PO bid for 14 days 2. Regimen B: ceftriaxone 250 mg IM, or cefoxitin 2 g IM plus probenicid 1 gm PO, or a third generation cephalosporin, and doxycycline 100 mg PO bid for 14 days Safe sex instructions* 13/88 (15%) Follow-up* 21/88 (24%) Return instructions* 61/88 (69%)

GC = Gonorrhea; Ch = Chlamydia; b-hCG = human chorionic gonadotropin; PID = pelvic inflammatory disease. *Components used for selected criteria.

For patients in the PID group, only those who were discharged from the ED with the diagnosis of PID were included in the analysis of antibiotic use and discharge instructions. Antibiotic choices for admitted patients with PID are often controlled by the admitting service; therefore, these patients were eliminated from the analysis. This subset of patients would also not have discharge instructions. Data were analyzed by using SPSS version 11.0 (SPSS Inc., Chicago, IL). Descriptive statistics including means and standard deviations for continuous variables are reported. All other data are presented as percentages. Categorical data were compared by using

chi-square analysis. The p-values less than 0.05 were considered statistically significant.

RESULTS By using the ICD-9 codes, 246 patients were identified within the study time period. Of these, 203 (85%) were enrolled. The remaining 43 (17%) were excluded for the following reasons: 24 (56%) did not meet age criteria, 8 (19%) were missing documentation of the ED visit in the medical record, and 11 (26%) medical records were not available. Table 2 provides demographics and proportions of patients with the

374

Kane et al.

TABLE 2. Demographics of the Study Population

For gonorrhea, 20 of 48 (42%) men and 13 of 34 (38%) women received appropriate treatment. The most frequently used drug was ceftriaxone, used at the proper dose of 125 mg intramuscularly (IM) in 19 of 48 (40%) men and 13 of 34 (38%) women, but at incorrect doses of 250 mg IM in 18 of 48 (38%) and 11 of 34 (32%) patients, respectively. Even allowing for an increase in dosage of ceftriaxone from 125 mg to 250 mg in patients with urethritis and cervicitis, the rates of compliance increased only to slightly more than half, or 58%. Chlamydia was treated with recommended drugs in 35 of 48 (73%) men and 21 of 34 (62%) women with 1-g single-dose azithromycin therapy. Of the 121 women in the PID group, 88 (73%) were discharged to home. Of these 88 discharged patients, only 28 (32%) received one of the two CDC-recommended outpatient antibiotic regimens. The majority, 85%, received Regimen B, which is ceftriaxone and doxycycline. Common errors in antibiotic selection for PID included the administration of single-dose therapy of azithromycin in 21 of 60 (35%) patients and less than the recommended 14 days of doxycycline therapy in 13 of 60 (22%) patients. At the time of discharge, the CDC recommends treatment for partners and the abstinence from sexual activity for all three diseases. Safe sex instructions were documented for 24 of 48 (50%) of the urethritis group, 6 of 34 (18%) of the cervicitis group, and 13 of 88 (15%) of the PID patients discharged to home. Safe sex instructions were documented significantly more often for men than women at time of discharge (50% for men; 16% for women; p # 0.001). These were the only required discharge instructions for urethritis and cervicitis. For PID, the CDC recommends follow-up assessment within 72 hours. A follow-up appointment was recommended for 21 of 88 (24%) of the women, and 61 of 88 (69%) were given instructions regarding their disease and reasons to seek medical care before their scheduled appointment. The overall compliance with the recommended discharge instructions for the PID group was 4 of 88 (3%). Complete compliance within each domain for each diagnosis for both the comprehensive criteria and the selected criteria is depicted in Table 3. For example, in 121 patients diagnosed with PID, 17 (14%) had documentation supporting all five comprehensive (full) criteria pertaining to the history domain for PID. This increased to 71 (59%) when selected criteria were used. However, there was little change when full or selected criteria were applied to the domain of physical examination (i.e., 27 [22%] compared with 34 [28%]). Total (100%) compliance for all five domains is extremely low: 3 of 48 (6%) for urethritis, 1 of 34 (3%) for cervicitis, and 0 (0%) for PID. Despite the increased compliance in one or two domains when the selected criteria were applied, the overall total compliance did not significantly change.

Characteristics Mean age in years (6SD) Gender n (%) Diagnosis n (%)

Total (N = 203)

Male (n = 48)

Female (n = 155)

25.8 (610.0) 203 Urethritis Cervicitis PID Total Admitted Discharged

28.5 (612.9) 48 (24) 48 (24) — —

24.8 (68.9) 155 (76) — 34 (22) 121 (60) 33 (16) 88 (43)

PID = pelvic inflammatory disease.

diagnosis of urethritis 48 (24%), cervicitis 34 (22%), and PID 121 (60%). The components of each domain for the comprehensive criteria are listed in Table 1. The number and proportion of time that each item was noted in the record are provided. The items used to create the subset of selected criteria are denoted by an asterisk. Certain identified deficits are particularly striking, such as the fact that only 39 of 121 (32%) of cases of PID mentioned the presence or absence of adnexal tenderness, and only about three fourths of charts mentioned cervical motion tenderness, the hallmark of PID. All patients in the study should have had gonorrheal and chlamydial cultures sent. This is the only required diagnostic testing for the urethritis group and was done 79% of the time (38 of 48). The cervicitis and PID groups had GC/Ch cultures sent 77% of the time (29 of 34 and 93 of 121, respectively). There was statistically no significant difference in the likelihood of having a culture sent based on gender. For the female patients, diagnostic testing also included having a documented urine b-hCG. Twenty-nine of 34 (85%) of the cervicitis patients and 105 of 121 (87%) of patients with the diagnosis of PID had this information recorded. The overall diagnostic compliance for females was 24 of 34 (71%) for cervicitis and 86 of 121 (71%) for PID. The most important domains for quality measurements are treatment and discharge instructions. Unfortunately, recommended antibiotic regimens for gonorrhea and chlamydia were prescribed correctly less than one third of the time. There was no statistically significant difference in the likelihood of receiving CDC-recommended antibiotics at the time of discharge based on gender. Urethritis and cervicitis have the same antibiotic recommendations according to the CDC. Both require treatment for gonorrhea and chlamydia with separate drugs. The urethritis group received recommended treatment for both pathogens 33% of the time (16 of 48), and the cervicitis group 32% of the time (11 of 34).

d

COMPLIANCE WITH CDC FOR STD TREATMENT

ACAD EMERG MED

d

April 2004, Vol. 11, No. 4

d

375

www.aemj.org

TABLE 3. Compliance with Examination, Testing, and Treatment: Comprehensive (Full) vs. Selected Criteria Urethritis (n = 48)

History Physical examination Diagnostic testing Compliance with antibiotics for discharged patients Discharge instructions Total

Cervicitis (n = 34)

PID (n = 121)

Full Criteria n (%)

Selected Criteria n (%)

Full Criteria n (%)

Selected Criteria n (%)

Full Criteria n (%)

Selected Criteria n (%)

35 (73) 30 (63) 38 (79)

46 (96) — —

5 (15) 5 (15) 24 (71)

14 (41) 11 (32) —

17 (14) 27 (22) 86 (71)

71 (59) 34 (28) —

16 (33) 24 (50) 3 (6)

28 (58) — 8 (17)

11 (32) 6 (18) 1 (3)

20 (58) — 1 (3)

28/88* (32) 4/88* (3) 0 (0)

— — 0 (0)

*Discharged with pelvic inflammatory disease (PID) (n = 88).

DISCUSSION This study revealed that the overall compliance with the CDC recommendations for the diagnosis and treatment of STDs measured by documentation is extremely poor. The results were similar whether strict comprehensive criteria were used or a less stringent subset of selected criteria was used. One could argue which subsets of items in each domain are essential, but adding or subtracting items would not significantly change the results. There are several hypotheses as to why compliance is so low. Successful implementation of clinical guidelines has been reported to require a four-step process.14 This includes the development of evidence-based guidelines, dissemination and discussion, implementation with feedback, and practitioner accountability. A formal feedback and accountability system regarding treatment of STDs has not been formally developed at this institution. Continuing quality improvement measures are often used when triggered by complaints, unscheduled return visits, or positive culture results. It is conceivable that compliance was poor due to the fact that this study was performed in a teaching institution with multiple rotating resident physicians who may not have been exposed to the guidelines. However, every patient was seen by an attending physician who theoretically should have examined the patient and been involved in antibiotic choices. It is possible that the physical examination was completed and not documented because the study relied on review of medical records, which may suggest that deficits identified are due to documentation and not quality. However, this issue has been studied by comparing process-based quality scores using chart extraction with the use of vignettes and standardized patients. Peabody et al.15 reported only an approximately 5% and 10% lowering of process scores when comparing chart abstraction with the use of vignettes and standardized patients, respectively. Another study reported a 6.4% false-positive rate in medical record documentation compared with structured reports by standardized patients.16 Therefore, at most, our scores would have been approximately 10%

higher if another method had been used to collect data. The largest measure of compliance in the study was documentation on the free-form medical record. A prompted, templated record specific to STDs may improve both patient care and documentation. Feder et al.17 found this type of templating to improve documentation and, presumably, patient care. Most concerning was the poor compliance with the recommended antibiotics, because once the provider has arrived at a diagnosis, the recommendations are easily accessible. This poor compliance in the area of antibiotic use may represent a low rate of practitioner familiarity with the recommendations. In an era in which drug resistance has become epidemic, proper use of antibiotics must be a priority for all providers. Even allowing for an increased dose of 250 mg of ceftriaxone instead of the recommended 125 mg, the antibiotic compliance was only 58% for urethritis and cervicitis. From a public health perspective, compliance with the diagnosis and treatment of STDs is crucial because the incidence has reached epidemic proportions.18 Untreated, the sequelae for women are significant and include progression of disease, impaired fertility or infertility, and increased risk of ectopic pregnancy. The consequence of untreated or improperly treated STDs in men is the continued risk for the transmission of disease. Prior surveys of practitioners related to the guidelines demonstrated that less than half of primary care physicians knew about or followed the recommendations for PID and found poor compliance with disposition criteria for adolescents diagnosed with PID.19 Hessol et al.,20 in a survey of primary care physicians, found only 44% reported following the CDC guidelines and that 52% were unaware of the guidelines or they did not follow them. One retrospective study looking at patients with documented positive cultures for gonorrhea or chlamydia found only 53% of patients were correctly diagnosed and treated in the ED.21 This current study differs from the above-mentioned studies in that the measurement used was the

376

Kane et al.

documentation of practitioner practice and not perceived practice. This may explain why the compliance rates are even lower than those previously reported. Although this is a study of one institution, it should not lure physicians into thinking that these results would not apply to them. Repeatedly, research has demonstrated that physicians have failed to comply with well-established guidelines, with an important example being poor compliance with prescribing beta-blockers and aspirin after myocardial infarction.22 Studies of both residents and practicing physicians have overestimated their adherence to guidelines with other conditions, including hypertension,23 lipid management,24 and diabetes management.25 Phillips et al.26 reported a phenomenon described as ‘‘clinical inertia,’’ defined as recognizing a problem but failing to act. These authors report that one of the major factors contributing to clinical inertia is that providers often overestimate the care they provide and that system changes must be implemented for change to occur. The public health issues would also seem to compel providers to comply with the guidelines, especially in terms of patient discharge instructions. This study did not try to track whether reporting of the diseases to the health department occurred, as required by state law in Connecticut, but the study did find a statistically significant difference in the number of men who received instructions compared with the number of women. Because women suffer the greatest morbidity and potential mortality related to STDs, the prevention of re-infection is of even greater importance for them, and this low percentage of safe sex instructions is particularly troublesome. One may postulate that this is related to gender bias and that the physician is uncomfortable in discussing STDs with women, particularly if the diagnosis is unclear. However, this study highlights the need for physician awareness of the importance of such behavior. The CDC in 2000 reported that STDs, especially chlamydial infections, are on the rise. This epidemic has prompted at least one state to consider giving patients medications to take home to treat their partners.27 Furthermore, at the time of the 1998 CDC guidelines dissemination, PID outpatient treatment regimens for PID had never been compared with inpatient treatment. In 2002, Ness et al. reported the results of the PEACH study (n = 831).28 The authors concluded that women with mild to moderate PID can be safely treated as outpatients; however, every patient received CDCrecommended treatment, discharge instructions, and follow-up. At about the same time, the CDC simplified their diagnostic criteria for the diagnosis of PID. However, the guidelines call for the discretionary addition of Flagyl to outpatient treatment, which may add complexity to the discharge instructions.29 Given the serious morbidity related to PID and the potential for outpatient treatment failure, which the CDC lists as

an indication for admission, the low rate of documentation of discharge instructions and lack of follow-up appointments might suggest a return to admission for all cases of PID. In addition, the International-Infectious Disease Society for Obstetrics and Gynecology-USA wrote in 2001 that it advocated greater hospitalization for PID than the CDC recommends.30 Most studies have demonstrated that specific interventions are required to improve physician compliance with practice guidelines. Continuing education without administrative initiatives such as standard order sheets, standardized charting, or computer-based interventions has not been shown to be effective.31,32 Studies on improving quality care and implementing change show that certain strategies such as emergence of opinion leaders, individual feedback, and system changes including computer-assisted decision support, automated data entry, and retrieval for measurements are effective in improving compliance.33 The deficits in quality care identified in this study are substantial. Although one may argue over the magnitude of the problem, a significant gap exists between what we know works and what is actually done, leading to potentially serious morbidity. Future areas of study should include focused practitioner education, standardized discharge instructions, and prompts in the medical record, the computerized diagnostic ordering system, and pharmacy substations. This requires a systems-based approach to improve patient care to change physician practice.34,35

d

COMPLIANCE WITH CDC FOR STD TREATMENT

LIMITATIONS One limitation of the study is that it was conducted at only one teaching institution, limiting the generalizability of the results. A retrospective chart review, using documentation as a measure of compliance, represents another limitation of this study. However, the measurement of compliance in diagnostic testing, specifically culturing, allowed for the use of institution-required computer order entry of the test as well as the laboratory documentation of the test results. These supplementary sources of documentation, independent of the medical records completed by the providers, may explain why the compliance rates were highest for diagnostic testing. However, as stated above, research demonstrates that other methods of data collection may have raised our compliance rates, at most, approximately 10%. It is most likely that the overall compliance for cervicitis and PID are lower than urethritis because the number of items necessary in each domain is greater. However, this did not change even with a more select subset of criteria. The selection of the subset is, however, artificial. The comprehensive guidelines are the official recommendations of the CDC, and perhaps we should expend more energy in determining how to make individual and system changes that increase

www.aemj.org

377

compliance with these recommendations than in how to arbitrarily choose the ‘‘essential’’ ones.

16. Dresselhaus TR, Luck J, Peabody JW. The ethical problem of false positives: a prospective evaluation of physician reporting in the medical record. J Med Ethics. 2002; 28:291–4. 17. Feder G, Griffiths C, Highton C, Eldridge S, Spence M, Southgate L. Do clinical guidelines introduced with practice based education improve care of asthmatic and diabetic patients? A randomized controlled trial in general practices in east London. BMJ. 1995; 311:1473–8. 18. Centers for Disease Control and Prevention. Tracking the hidden epidemics: trends in STDs in the United States. 2000. Available at: http://www.cdc.gov/nchstp/dstd/ Stats_Trends/Stats_and_Trends.htm. Accessed Feb 2, 2004. 19. Benaim J, Pulaski M, Coupey SM. Adolescent girls and pelvic inflammatory disease. Arch Pediatr Adolesc Med. 1998; 152:449–54. 20. Hessol NA, Priddy FH, Bolan G, et al. Management of pelvic inflammatory disease by primary care physicians. A comparison with Centers for Disease Control and Prevention guidelines. Sex Transm Dis. 1996; 23:157–63. 21. Yealy DM, Greene TJ, Hobbs GD. Underrecognition of cervical Neisseria gonorrhoeae and Chlamydia trachomatis infections in the emergency department. Acad Emerg Med. 1997; 4:962–7. 22. Ellerbeck EF, Jencks SF, Radford MJ, et al. Quality of care for Medicare patients with acute myocardial infarction. JAMA. 1995; 273:1509–14. 23. Hyman DJ, Pavlik VN. Self-reported hypertension treatment practices among primary care physicians: blood pressure thresholds, drug choices, and the role of guidelines and evidence-based medicine. Arch Intern Med. 2000; 160:2281–6. 24. McBride P, Schrott HG, Plane MB, Underbakke G, Brown RL. Primary care practice adherence to National Cholesterol Education Program guidelines for patients with coronary heart disease. Arch Intern Med. 1998; 158:1238–44. 25. Drass J, Kell S, Osborn M, et al. Diabetes care for Medicare beneficiaries. Attitudes and behaviors of primary care physicians. Diabetes Care. 1998; 21:1282–7. 26. Phillips LS, Branch WT, Cook CB, et al. Clinical inertia. Ann Intern Med. 2001; 135:825–34. 27. Elliott VS. Extra meds for sex partners: one approach in public health struggle to curb Chlaymdia. Am Med News. 2001; 44:1–2. 28. Ness RB, Soper DE, Holley RL, et al. Effectiveness of inpatient and outpatient treatment strategies for women with pelvic inflammatory disease: results from the pelvic inflammatory disease evaluation and clinical health (PEACH) randomized trial. Am J Obstet Gynecol. 2002; 186:929–37. 29. Centers for Disease Control and Prevention. Sexually transmitted diseases: treatment guidelines 2002. MMWR. 2002; 51(RR-6):48–50, 1. 30. Hemsel DL, Ledger WJ, Martens M, Monif GRG, Osborne NG, Thomason JL. Concerns regarding the Centers for Disease Control’s published guidelines for pelvic inflammatory disease. Clin Infect Dis. 2001; 32:103–7. 31. Davis DA, Thomson MA. Changing physician performance. A systematic review of the effect of continuing medical education strategies. JAMA. 1995; 274:700–5. 32. Davis DM, O’Brien MA. Impact of formal continuing medical education: do conferences, workshops, rounds, and other traditional continuing education activities change physician behavior or health care outcomes? JAMA. 1999; 282:867–74. 33. Grol R. Improving the quality of medical care: building bridges among professional pride, payer profit, and patient satisfaction. JAMA. 2001; 286:2578–85. 34. Greco PJ, Eisenberg JM. Changing physicians’ practices. N Engl J Med. 1993; 329:1271–3. 35. Lundberg GD. Changing physician behavior in ordering diagnostic tests [editorial]. JAMA. 1998; 280:2036.

ACAD EMERG MED

d

April 2004, Vol. 11, No. 4

d

CONCLUSIONS This study demonstrates poor compliance rates of ED practitioners with a set of national, generally accepted, clinical guidelines. Poor compliance with antibiotic recommendations and discharge instructions raise questions concerning the safe discharge from the ED of women diagnosed with PID. Finally, despite the growing STD epidemic, safe sex instructions are provided to women significantly less frequently than to men. Strategies to improve these deficits in quality care need to be developed. References 1. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med. 2003; 348:2635–45. 2. Durieux P, Nizard R, Ravaud P, Mounier N, Lepage E. A clinical decision support system for prevention of venous thromboembolism: effect on physician behavior. JAMA. 2000; 283:2816–21. 3. Lobach DF, Hammond E. Computerized decision support based on a clinical practice guideline improves compliance with care standards. Am J Med. 1997; 102:89–98. 4. Rothschild JM, Lee TH, Bae T, Bates DW. Clinician use of a palmtop drug reference guide. JAMA. 2002; 9:223–9. 5. Christakis DA, Zimmerman FJ, Wright JA, Garrison MM, Rivara FP, Davis RL. A randomized controlled trial of pointof-care evidence to improve the antibiotic prescribing practices for otitis media in children. Pediatrics. 2001; 107:E15. 6. Grimshaw JM, Russell IT. Effect of clinical guidelines on medical practice: a systematic review of rigorous evaluations. Lancet. 1993; 342:1317–22. 7. Cline DM, Welch KJ, Cline LS, Brown CK. Physician compliance with advanced cardiac life support guidelines. Ann Emerg Med. 1995; 25:52–7. 8. Hardern RD, Hampshaw S. What do accident and emergency medical staff think of practice guidelines? Eur J Emerg Med. 1997; 4:68–71. 9. Dean B, Schacter M, Vincent C, Barber N. Causes of prescribing errors in hospital inpatients: a prospective study. Lancet. 2002; 359:1373–8. 10. Lewis LM, Lasater LC, Ruoff BE. Failure of a chest pain clinical policy to modify physician evaluation and management. Ann Emerg Med. 1995; 25:9–14. 11. Lomas J, Anderson GM, Domnick-Pierre K, Vayda E, Enkin MW, Hannah WJ. Do practice guidelines guide practice? The effect of a consensus statement on the practice of physicians. N Engl J Med. 1989; 321:1306–11. 12. Huang ES, Stafford RS. National patterns in the treatment of urinary tract infections in women by ambulatory care physicians. Arch Intern Med. 2002; 162:41–7. 13. Centers for Disease Control and Prevention. 1998 Guidelines for treatment of sexually transmitted diseases. MMWR. 1998; 47(RR-1):49–52, 59-62 for urethritis; 52-56, 59-62 for cervicitis; 79-85 for PID. 14. Smith TJ, Hillner BE. Ensuring quality cancer care by the use of clinical practice guidelines and critical pathways. J Clin Oncol. 2001; 19:2886–97. 15. Peabody JW, Luck J, Glassman P, Dresselhaus TR, Lee M. Comparison of vignettes, standardized patients and chart abstraction. JAMA. 2000; 283:1715–22.