Preventive Care in the Emergency Department ... - Wiley Online Library

1042

PREVENTIVE CARE

Irvin et al. • PREVENTIVE SERVICES EVIDENCE-BASED REVIEW

Preventive Care in the Emergency Department, Part II: Clinical Preventive Services—An Emergency Medicine Evidence-based Review* CHARLENE BABCOCK IRVIN, MD, PETER C. WYER, MD, LOWELL W. GERSON, PHD, FOR THE SOCIETY FOR ACADEMIC EMERGENCY MEDICINE PUBLIC HEALTH AND EDUCATION TASK FORCE PREVENTIVE SERVICES WORK GROUP†

Abstract. Introduction: Emergency departments (EDs) provide an opportunity to initiate preventive services for millions of Americans who have no other source for these services. Objectives: To identify primary and secondary preventive interventions appropriate for inclusion in routine emergency care and, secondarily, to recommend areas in which research into the efficacy and cost–effectiveness of interventions is needed. Methods: Systematic reviews were performed on 17 candidate preventive interventions with potential applicability in the ED. All but one was selected from those reviewed by the U.S. Preventive Services Task Force (USPSTF). Each two-person review team followed a template that provided a uniform approach to search strategy, selection criteria, methodology appraisal, and analysis of the results of primary studies bearing on ED cost–effectiveness. Assigned proctors provided methodological guidance to the review teams throughout the review process. A grading scheme was developed that took into account the evidence and recommendations of the USPSTF supporting primary efficacy of the intervention and the level of evidence supporting ED application identified by the Society for Academic Emergency Medicine Public Health and Education Task Force (PHTF) review teams. Results: Seventeen reviews were completed. The following interventions received an alpha rating, indicating that evidence is sufficient to support offering these services in the ED setting, assuming sufficient resources are available: alcohol screening and intervention, HIV screening and referral (in high-risk, high-prevalence populations), hypertension

From the Department of Emergency Medicine, St. John Hospital and Medical Center, and Department of Emergency Medicine, Wayne State University School of Medicine, Detroit, MI (CBI); Columbia University College of Physicians and Surgeons, New York, NY (PCW); and the Division of Community Health Sciences, Northeastern Ohio Universities College of Medicine, Rootstown, OH (LWG). Received: April 21, 2000; accepted April 23, 2000. *Part I appears earlier in this issue. † Charlene B. Irvin, MD, Project Director, Template Co-de-

screening and referral, adult pneumococcal immunizations (age ⱖ65 years), referral of children without primary care physicians to a continuing source of care, and smoking cessation counseling. Interventions receiving a beta or gamma rating, indicating that existing research is not sufficient to recommend for or against instituting them routinely in the ED, include: identification and counseling of geriatric patients at risk of falls, Pap tests in women having a pelvic exam in the ED, counseling for smoke detector use, routine social service screening, depression screening, domestic violence screening, safe firearm storage counseling, motorcycle helmet use counseling, and youth violence counseling programs in the ED. Interventions not recommended for ED implementation (omega rating) include Pap test screening for women not having a routine pelvic exam, diabetes screening, and pediatric immunizations. Conclusions: A set of recommendations for prevention, screening, and counseling activities in the ED based on systematic reviews of selected interventions is presented. The applicability of these primary and secondary preventive services will vary with the different clinical environments and resources available in EDs. The PHTF recommendations should not be used as the basis of curtailing currently available services. This review makes clear the need for further research in this important area. Key words: prevention; counseling; screening; emergency services; review; recommendations. ACADEMIC EMERGENCY MEDICINE 2000; 7:1042–1054

signer, Topic Reviewer (HIV); Peter C. Wyer, MD, Template Designer and Proctor; Lowell W. Gerson, PhD, Template Codesigner, Proctor, and Topic Reviewer (Geriatric Falls); Karin Rhodes, MD, Project Co-initiator and Topic Reviewer (Social Services); James A. Gordon, MD, MPA, Project Co-initiator and Topic Reviewer (Social Services); Robert A. Lowe, MD, MPH, Project Co-initiator, Template Co-designer. Topic Reviewers (in topic alphabetical order): Cardiovascular: David Kerr, MD, and Trisha Nouhan, MD; Child Abuse: Allen Walker, MD, and Scott Krugman, MD; Domestic Violence: Deirdre Anglin, MD,

ACADEMIC EMERGENCY MEDICINE • September 2000, Volume 7, Number 9

T

HE National Hospital Ambulatory Medical Care Survey1 estimates there were 94.4 million visits made to U.S. emergency departments (EDs) in 1997. Emergency departments are organized to provide emergent and urgent care, and many ED patients have high rates of chronic disease and injury risks.2 The ED may provide an important opportunity to initiate prevention, screening, and counseling services for millions of Americans who have no other source for these services.3 Providing preventive services in the ED may constitute an important contribution of emergency medicine (EM) to the comprehensive health care interests of society at large. The Society for Academic Emergency Medicine Public Health and Education Task Force (PHTF) initiated an evidence-based approach to evaluate preventive interventions most likely to be cost-effective options for application to ED patient populations. This report describes the review process and provides a summary of recommendations for prevention, screening, and counseling services in the ED.

1043

Preventive interventions that had possible applicability in the ED were evaluated using a template-mediated review process based on the established methodology of systematic reviews.4–6 Several initial assumptions were made. These included the need for outcome information on the efficacy of any intervention to be recommended for implementation, the understanding that most interventions will be dependent on resources available at any particular ED, and the continued importance of advocating for access to primary care so that comprehensive care could be secured. In order to focus on interventions with established efficacy, candidate interventions were selected from those considered and evaluated by the U.S. Preventive Services Task Force (USPSTF).7 Appendix A provides excerpts from the USPSTF Guide to Clinical Preventive Services that pertain to assumptions adopted by the PHTF in the course of performing its own reviews in these topic areas. Social service activities were also included because these are commonly performed in the ED and are

related to preventive care. An integrated surveillance system has important bearing on preventive care in the ED but also has implications that go far beyond the context of emergency care per se. A narrative review of the potential role of EDs in organized surveillance systems was performed as part of this PHTF project. As the USPSTF focused on primary and secondary preventive initiatives, so did this project. Because the ED setting is distinct from those found in primary care offices, not everything that is effective in primary care will be cost-effective or feasible within an ED. Selection of candidate intervention incorporated the assumption that different EDs have different patient populations and hospital resources. Interventions that were clearly impractical in an ED setting, such as screening colonoscopy for colorectal cancer, were excluded. Team leaders from the PHTF selected general categories addressed by the USPSTF as areas for possible review.3 These included general issues such as vaccination or specific populations such as elder patients as general target areas. Initial reviewers for each general area were selected from the PHTF members. Each initial reviewer was asked to select a partner or partners with interests in the general topic area. Partners were not required to be members of the PHTF. However, all participants received the template and instructional materials. Topic reviewers selected at least one specific candidate preventive intervention from those included within the general category. The review team’s selection of potential candidate interventions was reviewed by the leadership team to ensure that they were consistent with the project’s objectives and that there was no overlap between the topics selected. The leadership team created a template that incorporated the methodology of systematic reviews and that also took into account the special features of the PHTF project, including its relationship to the USPSTF effort (Appendix A). At the outset of the structured review, all team members were sent a packet of materials that included the template, a detailed explanation of the template, worksheets for the individual study appraisals, the scoring system, and a cover note with general instructions. No other formal instruction in the methodology of sys-

MPH, and Carolyn Sachs, MD, MPH; Geriatric Screening: Lowell Gerson, PhD, and John V. Weigand, MD; Immunizations—Adult: Robert D. Powers, MD, MPH, and Stuart Brilliant, MD; Injury Prevention: Herbert Garrison, MD, MPH, Leslie Maas, MHS, Stephen Hargarten, MD, MPH, Mallory O’Brian, MS, and Guohua Li, MD; Mental Health: Ken Parsons, MD, and Terry Kowalenko, MD; Neoplastic: Kimberly Collins, MD, and Denise Nassisi, MD; Pediatrics: Lori Montagna, MD, and Bruce Becker, MD, MPH; Social Services: James A. Gordon, MD, MPA, and Karin Rhodes, MD; Sexually

Transmitted Diseases: Richard Rothman, MD, PhD, Theresa Dolan, MD, Charlene Irvin, MD, and David Slobodkin, MD, MPH; Substance Abuse: Gail D’Onofrio, MD, and Linda Degutis, DrPH; Emergency Department Surveillance: Jon Mark Hirshon, MD, MPH; Tobacco: Steve Bernstein, MD, and Bruce Becker, MD, MPH; Youth Violence: Elizabeth Datner, MD, and Kimberly Collins, MD. Address for correspondence and reprints: Charlene B. Irvin, MD, 49884 Miller Court, Chesterfield, MI 48047. Fax: 810-7161703; e-mail: [email protected]

METHODS

1044

PREVENTIVE CARE

tematic reviews was provided to the review teams. In order to compensate for this possible limitation, a proctorship system was set up to help steer teams through a potentially unfamiliar process. The candidate interventions were divided arbitrarily into two groups and a proctor was assigned to the review teams within each group. The proctor served as an ongoing ‘‘mentor,’’ reviewer, and methodological advisor to the review efforts and editor of the topic write-ups. Based on early results, a somewhat revised template and worksheet were distributed to the teams several months after initiation of the reviews. The revisions were clarifications of the original methods. The template involved four phases. In the first phase, the USPSTF definition of the candidate intervention and the primary and secondary outcome measures considered by the USPSTF for this intervention were identified. In the second phase, the USPSTF recommendations (Appendix B) and the principal studies cited by the USPSTF in support of their recommendations were reviewed as the point of departure for the PHTF review of the efficacy and applicability of the same interventions in the ED setting. The third phase involved a structured literature search, selection of studies bearing on the issue of ED implementation, and methodological assessment of the selected studies. Selection and appraisal were performed independently by at least two reviewers. Reviewers reported the number of inconsistencies in their assessment and discussed within the team any differences. The proctor and leadership team adjudicated any unresolved differences. In addition to outcome information, data bearing on the practicality and feasibility of implementation in ED settings were emphasized in the analyses of the individual studies. Data from the individual studies were abstracted onto separate worksheets. In the fourth and last phase of the templated review, data from all of the selected studies were considered. Conclusions regarding the probable effectiveness and applicability of the interventions in an ED population were formulated in relationship to the strength of evidence at hand. Literature searches routinely used MEDLINE and screened the bibliographies of all chosen articles. The template included suggested approaches to the MEDLINE searches. The use of available research librarian support was encouraged. The search protocol also encouraged the use of the Cochrane Library8 to identify relevant systematic reviews and controlled trials and of indexed searches of the abstracted and nonabstracted databases of Emergency Medical Abstracts9 to identify studies pertaining to ED implementation of the candidate interventions. Recommendations from the Advisory Committee on Immunization


Practices (ACIP),10 which appeared after the publication of the USPSTF recommendations, were used in determining the efficacy of the pneumococcal vaccine. A system for scoring methodological quality was developed for randomized trials, cohort studies, case–control studies, and studies on diagnostic performance of screening tests (Appendix C) using adaptations of published criteria.11,12 Review teams were instructed to first group selected studies by outcome and population and then to rank them by study design, using the USPSTF scheme, and by methodological quality score. When multiple studies were found in a resulting subcategory, the teams were instructed to give primary consideration to studies reflecting the highest two levels of study design. Statistical pooling of outcome results was not attempted. The project leadership attended a retreat after the initial reviews were completed and the template and worksheet forms were submitted. Each template was reviewed for consistency and completeness during this two-day period. Variability in the criteria being used by the review teams to formulate their recommendations was addressed. To achieve consistency and objectivity in the formulation of recommendations regarding the appropriateness of the candidate interventions for ED application, a grading system was adopted. This took into account both the level of recommendation and evidence offered by the USPSTF for that preventive service and the evidence independently identified by the PHTF review teams pertaining to the efficacy and feasibility of ED implementation of the same interventions. The grades were described as alpha, beta, gamma, and omega. Based on a systematic application of the principles embodied in this grading scheme, the candidate interventions were graded. There was consensus between the preventive review teams and the project leadership regarding the assigned grading scheme. The alpha, beta, gamma, and omega grading scheme was defined as follows: • Alpha: The USPSTF recommends this prevention, screening, or counseling intervention (A or B). Evidence is sufficient to support offering this in the ED setting, assuming sufficient resources are available. • Beta: The USPSTF recommends this prevention, screening, or counseling intervention (A or B). Existing research is not sufficient to recommend for or against offering this preventive service routinely in the ED. Research into this area is strongly encouraged. • Gamma: The USPSTF found insufficient evidence to make a recommendation on this prevention, screening, and counseling intervention (C). Evidence to recommend for or against offering this

1045


preventive service routinely in the ED is also insufficient. Research on the efficacy on interventions in this area is needed. • Omega: Regardless of the level of the USPSTF recommendation, this is not recommended for ED implementation.

RESULTS Table 1 lists the preventive categories considered, and the specific topic reviewed. There were 15 general categories considered, and 17 specific topics reviewed. Each team was required to review one specific topic in their general category. Some teams reviewed more than one topic. Table 2 presents the summary of the PHTF recommendations. In some cases, more than one graded recommendation, corresponding to different ED patient subgroups, was made with respect to a single candidate intervention. Based on outcome data and assuming adequate availability of resources, the following preventive interventions are recommended for direct application in the ED setting (alpha rating): alcohol screening and intervention, HIV screening and referral (in high-risk, high-prevalence populations), hypertension screening and referral, adult pneumococcal immunizations (age ⱖ65 years), referring children without primary care physicians to a continuing source of care, and smoking cessation counseling. There is not sufficient evidence from ED-based studies to recommend for or against identification and counseling of geriatric patients at risk of falls, Pap tests in women having a pelvic exam in the ED, counseling for smoke detector use, and social service screening as routine new preventive activities. The USPSTF recommendation for these candidate interventions was A or B. Social service screening was not reviewed by the USPSTF, but there is evidence in favor of recommending social services. These candidate interventions received a beta rating because they have been demonstrated to be effective in non-ED settings. It is likely that they would prove to be effective, if adequately studied. Research into these interventions is strongly encouraged. Depression screening, domestic violence screening, safe firearm storage counseling, child abuse screening, motorcycle helmet use counseling, and youth violence counseling cannot be unconditionally recommended for routing use in the ED setting (gamma rating). The USPSTF recommendation for them was C (insufficient evidence to recommend or discourage), and no compelling ED studies were found that would override the USPSTF assessment. Research into the primary efficacy of these candidate interventions is necessary.

TABLE 1. Preventive Intervention Catgories Considered Preventive Category Considered

Specific Preventive Intervention Reviewed

Cardiovascular

Hypertension screening Diabetes screening

Child abuse

Physical child abuse

Domestic violence

Domestic violence screening

Geriatric screening

Fall prevention screening

Immunization—adult

Pneumococcal immunization (age ⱖ65 yr)

Injury prevention

Safe storage of firearms Smoke detector use counseling Motorcycle helmet use counseling

Mental health

Depression screening

Neoplastic screening

Pap screening in women

Pediatrics

Pediatric immunizations

Social services

Primary care referral General social services

Sexually transmitted disease

HIV screening

Substance abuse

Alcohol abuse screening

Tobacco

Tobacco cessation counseling

ED surveillance

ED surveillance

Youth violence

Youth violence counseling

The following candidate interventions are not recommended to be performed in the ED setting regardless of the USPSTF recommendation (omega rating): Pap test screening on women not having a routine pelvic exam, diabetes screening, and pediatric immunizations. Detailed descriptions of specific topic reviews are planned as separate submissions.

DISCUSSION The PHTF approach to identifying potentially useful preventive interventions for ED implementation embodied several assumptions. One important assumption, inherent in the adoption of the USPSTF and the PHTF recommendations, is the importance of the concept of cost–effectiveness. Neither the USPSTF nor the PHTF performed formal cost–effectiveness analyses regarding the preventive interventions they considered. Nor did we find such analyses in the course of the vast majority of interventions they researched (Appendix A). Nonetheless, the concept of cost–effectiveness was

1046

TABLE 2. SAEM PHTF Summary Preventive Services Recommendations, June 2000 Candidate Intervention (Reviewers)

Ideal Outcome

Outcome Available from Existing Studies Reviewed

ED Population

Reduction in alcohol consumption in those at risk, decreased ED visits

All adolescents and adults considered at risk

Alpha

B

HIV‡ screening Richard Rothman, MD, PhD Theresa Dolan, MD Charlene Irvin, MD

Prevention of morbidity and mortality; early recognition and treatment

Early detection and successful referral

High-risk population‡ in areas with high prevalence§ of disease

Alpha

A for high-risk population

Hypertension David Kerr, MD Trisha Nouhan, MD

Prevention or minimization of mortality and morbidity due to CNS, renal, or cardiovascular disease

Successful early detection and referral

All adults

Alpha

A

Pneumococcal vaccination Robert Powers, MD, MPH Stuart Brilliant, MD

Reduction in invasive pneumococcal disease, adults >65 years

Increased vaccination rate

Age >65 years

Alpha

B‫ن‬

Smoking cessation counseling Steve Bernstein, MD Bruce Becker, MD, MPH

Reduction of tobacco-related morbidity and mortality

Smoking cessation

All smokers

Alpha

A

Social services needs assessment and referral James A. Gordon, MD, MPA Karin Rhodes, MD

Prevention of morbidity-related to poor social circumstance

Decreased ED visits and successful referral to primary care provider

Children

Simple referral to primary care provider Global needs assessment and referral to community resources

Alpha

Not graded

Beta

Not graded

Geriatric falls Lowell Gerson, PhD John Weigand, MD

Reduction of morbidity and mortality secondary to unintentional falls

Decreased number of falls

Age >64 years

No ED-based studies of secondary prevention; studies of falls management show multidimensional programs prevent falls and fractures

Beta

B

Pap smears Denise Nassisi, MD Kim Collins, MD

Reduction in mortality from cervical cancer

Cervical cancer detection rate

Sexually active women having ED pelvic exam Sexually active women not having ED pelvic exam All women in ED not up-to-date on Pap

Essential that follow-up procedure be established

Beta

A

Not feasible with usual resources in ED

Omega

Intervention limited to information pamphlet distribution

Beta

Referral for Pap smears

All

Screening suggested only in high-risk‡/high-prevalence§ population assuming pretest counseling resources available, institutional liability addressed, and established follow-up protocol available


Prevention of mortality and morbidity secondary to alcohol-related illnesses/injuries

USPSTF Grade†

PREVENTIVE CARE

Alcohol screening and intervention Gail D’Onofrio, MD Linda Degutis, DrPH

Special Considerations

SAEM PHTF Grade*

Presence of operational smoke detector

All adults

Ideal method of ED implementation not studied

Beta

B

Child abuse screening (physical) Allan Walker, MD Scott Krugman, MD

Reduction of death and disability in children resulting from physical child abuse

No screening studies available

All children

No ED-based studies evaluating primary and secondary preventive physical child abuse screening in the ED

Gamma

C

Depression screening Ken Parsons, MD Terry Kowalenko, MD

Suicide reduction

Successful screening

All adults

ED-based studies address screening only; early detection not shown to lead to improved outcome

Gamma

C

Domestic violence (DV) screening Deirdre Anglin, MD, MPH Carolyn Sachs, MD, MPH

Morbidity and mortality prevention from DV

Identification of DV victims in ED

All females >14 years

Impact of DV screening in ED unknown; routine screening effective at identification

Gamma

C

Firearm safe storage counseling Mallory O’Brien, MS Stephen Hargarten, MD, MPH

Decreased morbidity and mortality of firearm injuries

Removal or safe storage of home firearms

ED patients with firearm in home

No ED-based studies found

Gamma

B

Helmet, motorcycle usage counseling Herbert Garrison, MD, MPH Stephen Hargarten, MD, MPH

Morbidity or mortality reduction after crash

Increased observed helmet use

Motorcycle operators or riders

No ED-based studies found

Gamma

C

Violence, youth counseling Elizabeth Datner, MD Kimberly Collins, MD

Reduction in violence perpetration or victimization

Screening tool to identify youths at risk, decreased access to firearms

Parents of youths at risk, and youths at risk㛳

One screening tool studied (small study suggests benefit of injury prevention education)

Gamma

B, C

Diabetes screening David Kerr, MD Trisha Nouhan, MD

Prevention or minimization of mortality and morbidity related to diabetes

Hyperglycemia detection and referral

All patients

No ED screening studies; studies in clinic populations recommend not to routinely screen

Omega

C

Pediatric immunizations Bruce Becker, MD, MPH Lori Montagna, MD

Prevention of morbidity and mortality associated with lack of immunization

Increased immunizations rates

All children

Omega

A

Table footnotes continue on next page

1047

Reduction of morbidity and mortality from residential fires


Smoke detector counseling Leslie Maas, MHS Steve Hargarten, MD

Gamma: USPSTF found insufficient evidence to make a recommendation on this prevention, screening and counseling intervention (C). Evidence to recommend for or against offering this preventive service routinely in the ED is also insufficient. Research on the efficacy on interventions in this area is needed. Omega: Regardless of the level of the USPSTF recommendation, this is not recommended for ED implementation.

‡ High-risk population as defined as adolescent and adults: 1. Minority patients 25–44 years old. 2. Patients with heterosexual risk (ⱖ3 partners in last year, sex with a high-risk partner (includes past or current partner with HIV infection, bisexual partner, or partner with history of, or current, injection drug use), sex exchanged for money, multiple sexually transmitted diseases, currently seeking treatment for sexually transmitted diseases). 3. Past or present intravenous drug use. 4. Men who had sex with men after 1975. 5. Uninsured patients. 6. Pregnant women in populations where seroprevalence is >0.1%.

§ High prevalence as defined by CDC HIV seroprevalence rate of at least 1%, or an AIDS diagnosis rate >1 per 1,000 discharges.

‫ ن‬Advisory Committee on Immunization Practices recommendation 1997 ‘‘A.’’

㛳 Youths at risk = male, adolescent, African American, lower socioeconomic status, prior hospitalization for assault injuries, school problems or truancy, drug and/or alcohol use.


† The USPSTF grading system The following excerpt from Appendix A of the USPSTF Guide to Clinical Preventive Services, Second Edition, Report of the U.S. Preventive Services Task Force, U.S. Department of Health and Human Services, Office of Public Health and Science, Office of Disease Prevention and Health Promotion, 1996, summarizes the grading system used by the USPSTF in rating their strength of recommendations and the level of evidence found to support those recommendations. ‘‘Strength of Recommendations A: There is good evidence to support the recommendation that the condition be specifically considered in a periodic health examination. B: There is fair evidence to support the recommendation that the condition be specifically considered in a periodic health examination. C: There is insufficient evidence to recommend for or against the inclusion of the condition in a periodic health examination, but recommendations may be made on other grounds. D: There is fair evidence to support the recommendation that the condition be excluded from consideration in a periodic health examination. E: There is good evidence to support the recommendation that the condition be excluded from consideration in a periodic health examination.’’

PREVENTIVE CARE

* Public Health and Education Task Force (PHTF) recommendation definitions: Alpha: The U.S. Preventive Services Task Force (USPSTF) recommends this prevention, screening, or counseling intervention (A or B). Evidence is sufficient to support offering this in the ED setting, assuming sufficient resources are available. Beta: USPSTF recommends this prevention, screening, or counseling intervention (A or B). Existing research is not sufficient to recommend for or against offering this preventive service routinely in the ED. Research into this area is strongly encouraged.

1048

TABLE 2. Footnotes


an explicit premise of the USPSTF effort.7 Cost– effectiveness of a health care intervention, in principle, means that an additional, incremental outlay of health care resources (usually measured in units of monetary currency) results in a minimally acceptable additional increment of positive patient outcomes (measured in monetary or quality-of-life equivalents). There is a very important practical implication of this assumption. The PHTF addressed the issue of additional incremental allocations of resources, above those currently available in most, or in any particular, EDs. A beta or gamma recommendation for a particular intervention should be interpreted to mean that evidence from available clinical research does not justify an allocation of increased resources for this intervention. Reciprocally, such a recommendation from the PHTF does not necessarily mean that currently available and allocated resources for this service category, or for the human resources pertinent to this service, within any particular ED should be cut back or curtailed. Beta and gamma recommendations do illustrate the need for research. Beta recommendations focus on interventions shown to be effective elsewhere but untested in the ED. The research that is needed is to show that they can be performed effectively in the ED. Gamma recommendations indicate the existing level of evidence is such that studies testing the efficacy of the intervention in any primary care setting are needed. For example, domestic violence is an area dealing with a significant burden of suffering. Multiple organizations advocate for routine screening, such as the American College of Emergency Physicians and the American College of Obstetrics and Gynecology.13,14 To make a positive recommendation for this intervention, our review required quality data showing that available interventions and approaches reduce the burden of suffering and improve victim outcome. This level of evidence evaluating the effectiveness of domestic violence screening was not found for the emergency department setting. This literature review showing lack of evidence for the effectiveness of domestic violence screening should not be misinterpreted as evidence for lack of effectiveness of such screening. The next logical step is further research, not abandonment of attention to a problem with major public health impact. A second assumption made at the initiation of this project was the need for outcome information on the efficacy of any recommended intervention. Ideal outcome information for any preventive service is decreased morbidity and mortality from the disease or condition to which the preventive service was directed. However, the very nature of EM is that patients appear for episodic care, and fol-

1049

low-up of patients for long periods of time is extremely difficult. Understanding this, intermediate outcomes for ED studies were accepted in situations where the overall efficacy of the intervention had been demonstrated (USPSTF A-I recommendations). These intermediate outcomes included follow-up with a primary care provider for the first visit after a disease or condition was identified, or rates of successful immunization. We understand that most interventions recommended would be dependent on the availability of resources at any particular ED. In many cases, interventions that might be performed in the ED could consume significant resources. Although a particular intervention might have been assessed to fulfill the bottom-line cost–effectiveness threshold, the resources necessary for its successful implementation may not be available at all locations. The PHTF recommendations have been proposed with the understanding that implementation would be dependent on resource availability. Additionally, many of the candidate interventions would require ongoing primary care to follow up on the preventive initiative started. Successful ED initiation of these interventions is contingent on the direct availability of appropriate follow-up referrals. Primary and secondary prevention initiatives are not currently considered by many practitioners to be part of routine emergency care. Tertiary preventive interventions are generally acknowledged to be an inherent component of quality care within our specialty. Identifying the need for counseling and appropriate referral to those with direct evidence of the disease is within the acknowledged scope and practice of EM. Examples include identification of injured victims of child, elder, or spousal abuse, referral to alcohol detoxification centers for the chronic alcoholic patient who presents with a complication, or the uncontrolled hypertensive patient who has developed a hypertensive cardiomyopathy and presents with congestive heart failure. This project attempted to identify the evidence in support of primary and secondary prevention in the ED. It did not address tertiary preventive measures inherent to the immediate acute care of patients presenting to emergency departments. The USPSTF guide makes the distinction that for some health problems, clinical preventive services may not be as effective as community-level interventions. One example is youth and family violence where school-based and community programs, along with regulatory and legislative initiatives, might be more effective for preventing complications from these conditions than will preventive services delivered into the clinical setting. 7 We focused on clinical rather than commu-

1050

PREVENTIVE CARE

nity preventive services. As many community-level interventions may be linked to the ED, a beta or gamma recommendation for an intervention in the ED does not imply that the emergency department will not play a role in the community-level intervention. An important aspect of an evidence-based approach to a clinical area is the acknowledgment that the evidence itself is not constant. In this age of emphasis on clinical research, a review may be out of date by the time it is published. It will therefore be crucial to revisit and update the conclusions offered here as new research emerges in the coming years. This very methodology makes it a time-sensitive project. It is dependent on published research to date, and future published research may change these recommendations. By making explicit the process that we have used, we hope to provide a basis for emergency departments and programs to evaluate the relevance of our conclusions. By stating the limitations of available data, we also hope to inspire further research in many key areas.

LIMITATIONS The limitations of this project include the reliance on search strategies that may not have identified all relevant studies or studies currently in progress, limited information on cost or implementation strategies, topic reviewer’s limited experience in systematic reviews, and acceptance of surrogate outcomes in place of direct evidence of morbidity or mortality reduction. Another limitation is that stratification of recommendations based on outcomes may miss appropriate preventive interventions simply due to lack of relevant outcome studies. It is hoped that the beta and gamma recommendations will minimize this limitation. Finally, the USPSTF approach and recommendations were adopted as the starting point. The search strategies used by the USPSTF in preparing the guide, as well as other aspects of their approach, may ultimately be subject to challenge from the standpoint of the methodology of systematic reviews.


CONCLUSIONS A set of recommendations for prevention, screening, and counseling activities in the ED is presented. The applicability of the recommendations will vary with the different clinical environments and resources available in EDs. The PHTF recommendations should not be used as the basis of curtailing currently available services. This review makes clear the need for further research in this important area. References 1. Nourjah P. National Hospital Ambulatory Medical Care Survey: 1997 Emergency Department Summary. Advance data from Vital and Health Statistics; no. 304. Hyattsville, MD: National Center for Health Statistics, 1999. 2. Lowenstein SR, Koziol-McLain J, Thompson M, et al. Behavioral risk factors in emergency department patients: a multisite survey. Acad Emerg Med. 1988; 5:781–7 3. Rhodes KV, Gordon JA, Lowe RA, for the SAEM Public Health and Education Task Force Preventive Services Work Group. Preventive care in the emergency department, part I: Clinical preventive services—are they relevant to emergency medicine? Acad Emerg Med. 2000; 7:1036–41. 4. Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUORUM statement. Lancet. 1999; 354:1896–900. 5. The Cochrane Collaboration Handbook. The Cochrane Library, Version 3.0.2. Oxford, UK: Update Software, 1997. 6. Briss PA, Zaza S, Pappaioanou M, et al. Developing an evidence-based guide to community preventive services—methods. Am J Prev Med. 2000; 18(1 suppl):35–43. 7. U.S. Preventive Services Task Force. Guide to Clinical Preventive Services, Second Edition. Report of the U.S. Preventive Services Task Force, U.S. Department of Health and Human Services, Office of Public Health and Science, Office of Disease Prevention and Health Promotion, 1996. 8. The Cochrane Library, Issue 4. Oxford, UK: Update Software, 1999. 9. Monthly literature review and database published by the Center for Medical Education, Creamery, PA. 10. CDC. Prevention of pneumococcal disease: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR. 1997; 46(RR-08):1–24. 11. Jadad AR, Moore A, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Controlled Clin Trials. 1996; 17:1–12. 12. Guyatt GH, Rennie D. User’s guides to the medical literature [editorial]. JAMA. 1993; 270:2096–7. 13. American College of Emergency Physicians. Emergency medicine and domestic violence. Ann Emerg Med. 1995; 25: 442–3. 14. American College of Obstetrics and Gynecologists. ACOG technical bulletin no. 209: Domestic violence. Int J Gynaecol Obstet. 1995; 51:161–70.


1051

APPENDIX A Excerpts from the USPSTF Guide to Clinical Preventive Services, 2nd Edition*

The following excerpts from the Introduction are particularly relevant to the approach adopted by the SAEM Public Health and Education Task Force.

i. OVERVIEW ‘‘This report is intended for primary care clinicians: physicians, nurses, nurse practitioners, physician assistants, other allied health professionals, and students. It provides recommendations for clinical practice on preventive interventions—screening tests, counseling interventions, immunizations, and chemoprophylactic regimens—for the prevention of more than 80 target conditions. The patients for whom these services are recommended include asymptomatic individuals of all age groups and risk categoriesp.’’ (p. xxv) ‘‘For some health problems, community-level interventions may be more effective than clinical preventive services. Important health problems that are likely to require broader-based interventions than can be offered in the clinical setting alone include youth and family violence, initiation of tobacco use, unintended pregnancy in adolescents, and certain unintentional injuries.’’ (p. xxxii) A Research Agenda in Preventive Medicine ‘‘By reviewing comprehensively and critically the scientific evidence regarding clinical preventive services, the USPSTF identified important gaps in the literature and helped define targets for future clinical prevention research. Among the most important of these targets is more and better quality research evaluating the effectiveness of brief, directed counseling that can be delivered in the busy primary care practice setting. Given the importance of personal health practices, the scarcity of adequate evidence evaluating the effectiveness of brief counseling in the primary care setting is striking.’’ (pp. xxii–xxiii)

ii. METHODOLOGY ‘‘For many preventive services (and much of medical practice), there is insufficient evidence that the maneuver is or is not effective in improving outcomes (‘‘C’’ recommendation). This lack of evidence of effectiveness does not constitute evidence of ineffectiveness. A preventive service can lack evidence and receive a ‘‘C’’ recommendation because no effectiveness studies have been performed. In other cases, studies may have been performed but they may have produced conflicting results.

* U.S. Preventive Services Task Force. Guide to Clinical Preventive Services, Second Edition. Report of the U.S. Preventive Services Task Force, U.S. Department of Health and Human Services, Office of Public Health and Science, Office of Disease Prevention and Health Promotion, 1996.

Studies showing no benefit may lack adequate statistical power, making it unclear whether the maneuver would be proved effective if it were tested with a larger sample size. Studies showing a benefit may suffer from other design flaws (e.g., confounding variables) that raise questions about whether the observed effect was due to the experimental intervention or other factors. ‘‘In all of these instances, the Task Force gives the preventive service a ‘‘C’’ recommendation, noting that there is insufficient scientific evidence to conclude whether the maneuver should or should not be performed routinely. Practitioners and policy makers often need to consider factors other than science, however, in deciding how to proceed in the absence of evidence. The first of these considerations is potential harm to the patient.’’ (pp. li–lii)

v. COST –EFFECTIVENESS AND CLINICAL PREVENTIVE SERVICES ‘‘The fundamental role of effectiveness for clinical decisions motivated the U.S. Preventive Services Task Force effort to evaluate the evidence of effectiveness for clinical preventive services, and the Task Force recommendations in the following chapters are reflections of this evidence. ‘‘Effectiveness alone, however, is not a sufficient basis to initiate services in most practical health care contexts. Factors other than effectiveness, reflecting the immediate tradeoffs and broad implications of providing a service, are relevant to the goals and the practical constraints confronted by every decision maker.’’ (p. lxxxv) ‘‘Properly used, cost–effectiveness analysis incorporates and complements evidence of effectiveness to inform recommendations on clinical preventive and other health care services. It is intended not to substitute mechanically for complex decision-making processes but rather to be used in combination with other evidence. Efforts to enable cost–effectiveness analysis to be more easily, systematically, and usefully considered in policy decisions are under way.’’ (p. lxxxv) ‘‘Cost–effectiveness analysis is a method for assessing and summarizing the value of a medical technology, practice, or policy. Underlying the methodology is the assumption that the resources available to spend on health care are constrained, whether from the societal, organizational, practitioner, or patient point of view.’’ (p. lxxxvi) ‘‘The cost–effectiveness ratio summarizes information on cost and effect, allowing interventions to be compared on the basis of their worth and priority to the patient, society in general, or some other constituency. Although the cost–effectiveness ratio takes the form of a price—that is, a dollar cost per unit of effect—it is generally interpreted in the inverse manner, as a measure of the benefit achievable for a given level of resources.’’ (p. lxxxvi)

1052

PREVENTIVE CARE


APPENDIX B The USPSTF Grading System* The following excerpt from Appendix A summarizes the grading system used by the USPSTF in rating their strength of recommendations and the level of evidence found to support those recommendations. ‘‘Strength of Recommendations A: There is good evidence to support the recommendation that the condition be specifically considered in a periodic health examination. B: There is fair evidence to support the recommendation that the condition be specifically considered in a periodic health examination. C: There is insufficient evidence to recommend for or against the inclusion of the condition in a periodic health examination, but recommendations may be made on other grounds. D: There is fair evidence to support the recommendation that the condition be excluded from consideration in a periodic health examination. E: There is good evidence to support the recommendation that the condition be excluded from consideration in a periodic health examination. Determination of the quality of evidence (i.e., ‘good,’ ‘fair,’ ‘insufficient’) in the strength of recommendations

* U.S. Preventive Services Task Force. Guide to Clinical Preventive Services, Second Edition. Report of the U.S. Preventive Services Task Force, U.S. Department of Health and Human Services, Office of Public Health and Science, Office of Disease Prevention and Health Promotion, 1996.

was based on a systematic consideration of three criteria: the burden of suffering from the target condition, the characteristics of the intervention, and the effectiveness of the intervention as demonstrated in published clinical research. Effectiveness of the intervention received special emphasis. In reviewing clinical studies, the Task Force used strict criteria for selecting admissible evidence and placed emphasis on the quality of study designs. In grading the quality of evidence, the Task Force gave greater weight to those study designs that, for methodologic reasons, are less subject to bias and inferential error. The following rating system was used. Quality of Evidence I: Evidence obtained from at least one properly randomized controlled trial. II-1: Evidence obtained from well-designed controlled trials without randomization. II-2: Evidence obtained from well-designed cohort or case–control analytic studies, preferably from more than one center or research group. II-3: Evidence obtained from multiple time series with or without the intervention. Dramatic results in uncontrolled experiments (such as the results of the introduction of penicillin treatment in the 1940s) could also be regarded as this type of evidence. III: Opinions of respected authorities, based on clinical experience; descriptive studies and case reports; or reports of expert committees.’’ (pp. 861–862)

APPENDIX C SAEM PHTF Template; Supplement A, Quality Grading Scheme A: RANDOMIZED CONTROLLED TRIALS Question

Score (Points)

1. Was the study described as randomized?

1 for ‘‘yes’’ 0 for ‘‘no’’

2. Was the study described as double blind?


3. Was there a description of withdrawals and dropouts?


4. a) Was the method to generate the sequence of randomization described and was it appropriate (random numbers, computer generated, etc.)? b) Was it described but inappropriate (alternate allocation, by date of birth, chart number, etc.)?

1 0 ⫺1 0

for for for for

‘‘yes’’’ ‘‘no’’ ‘‘yes’’ ‘‘no’’

continued

1053


APPENDIX C (Cont.) SAEM PHTF Template; Supplement A, Quality Grading Scheme

A: RANDOMIZED CONTROLLED TRIALS Question 5. a) Was the method of double blinding described and appropriate (identical placebo, etc.)? b) Was it described but inappropiate (comparison of tablet with injection without double dummy, etc.)? 6. Was the loss to follow-up rate greater than 20%?

Score (Points) 1 0 ⫺1 0

for for for for

‘‘yes’’ ‘‘no’’ ‘‘yes’’ ‘‘no’’

⫺1 for ‘‘yes’’ 0 for ‘‘no’’

Maximum Score: 5 points Adapted from: Jadad AR, Moore A, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: Is blinding necessary? Controlled Clin Trials. 1996; 17:1–12.

B: COHORT STUDIES Question

Score

1. Was there a representative and well-defined sample of patients at a similar point in the course of the disease? (Were all subjects in the study appropriate targets of the intervention?)


2. Was follow-up sufficiently long and complete? (Was there adequate follow-up time for the outcome measures to be properly assessed?)


3. Was there adjustment for important prognostic factors? (Were factors other than the intervention controlled for?)


Maximum Score: 3 points Adapted from: Laupacis A, Wells G, Richardson WS, Tugwell P, for the Evidence-Based Medicine Working Group. Users’ guides to the medical literature. V. How to use an article about prognosis. JAMA. 1994; 272:234–237.

C: CASE –CONTROL AND ‘‘BEFORE/AFTER’’ STUDIES Question

Score

1. Were there clearly identified comparison groups that were similar with respect to important determinants of outcome, other than the one of interest?


2. Were the exposures and outcomes measured in the same way in the groups being compared?


3. Was follow-up sufficiently long and complete?


Maximum Score: 3 Adapted from: Levine M, Walter S, Lee H, et al., for the Evidence-Based Medicine Working Group. Users’ guides to the medical literature. IV. How to use an article about harm. JAMA. 1994; 271:1615–19. continued

1054

PREVENTIVE CARE


APPENDIX C (Cont.) SAEM PHTF Template; Supplement A, Quality Grading Scheme

D: STUDIES ON ACCURACY OF SCREENING TEST Question

Score

1. Was there an independent, blind comparison with a reference standard and was the reference standard appropriate?


2. Did the patient sample include an appropriate spectrum of patients to whom the diagnostic test will be applied in clinical practice? (Were the subjects in the study appropriate candidates for screening?)


3. Did the results of the test being evaluated influence the decision to perform the reference standard?


Maximum Score: 3 Adapted from: Jaeschke R, Guyatt GH, Sackett DL, for the Evidence-Based Medicine Working Group. User’s guides to the medical literature: III. How to use an article about a diagnostic test. A. Are the results of the study valid? JAMA. 1994; 271:389–91.

E: SIMPLE SURVEYS AND RETROSPECTIVE CHART REVIEWS* 1. If chart review was involved, were the methods of data extraction described and were their controls on the variability of interpretation of data from the medical record (such as inter-observer agreement measures)? 2. If a survey used a questionnaire, were the questions individually and the questionnaire as a whole subject to methods of validation? 3. Was the study performed at different treatment sites or have reports of similar studies been independently located? * No formal grading system was used for studies in this category.