The full content of Annals is available to subscribers

Subscribe/Learn More  >
Ideas and Opinions |

Candidate Performance Measures for Screening for, Assessing, and Treating Unhealthy Substance Use in Hospitals: Advocacy or Evidence-Based Practice?

Richard Saitz, MD, MPH
[+] Article, Author, and Disclosure Information

From Boston Medical Center, Boston University School of Medicine, and Boston University School of Public Health, Boston, Massachusetts.

Acknowledgment: The author thanks Dr. Jeffrey H. Samet and anonymous peer reviewers who critically reviewed the manuscript, which led to substantial improvements.

Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M09-2561.

Requests for Single Reprints: Richard Saitz, MD, MPH, Clinical Addiction Research and Education (CARE) Unit, Section of General Internal Medicine, Boston Medical Center, 801 Massachusetts Avenue, 2nd Floor, Boston, MA 02118; e-mail, rsaitz@bu.edu.

Ann Intern Med. 2010;153(1):40-43. doi:10.7326/0003-4819-153-1-201007060-00008
Text Size: A A A

This article has been corrected. For original version, click “Original Version (PDF)” in column 2.

The Joint Commission recently proposed candidate performance measures addressing unhealthy substance use in hospitalized patients. The proposed measures of screening and brief intervention (SBI) assume that interventions that work in one setting (primary care outpatient practice) would work in another (hospital); treatment would have the same benefits for persons identified by screening as for those with symptoms who seek help; treatments that work for persons less severely affected by substance use would also work for those with more severe illness; and an approach that works for nondependent, unhealthy alcohol use would work for drug use. However, these assumptions extrapolate evidence of the effectiveness of SBI for primary care outpatients with nondependent, unhealthy alcohol use to the inpatient setting, persons with dependence, and other substances. Although quality of care for unhealthy substance use in all medical settings needs to improve, the evidence base for SBI in the hospital is too limited for the implementation of performance measures assessing this care.





Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).


Submit a Comment/Letter
Targets and terror approach, insanity and improvement
Posted on July 15, 2010
Alain Braillon
GRES, Amiens. France
Conflict of Interest: None Declared

Unhealthy substance use in medical settings is a serious problem which needs improvement and Richard Saitz rightly challenged the indicators epidemics and pledged for evidence based quality.(1) This problem cannot justify for redisorganization.(2)

Most indicators are not relevant to clinical outcomes (morbidity/mortality) and evidence based indicators are still the Holly Graal. What happens when we extrapolate from evidence? Just look at the drug marketing authorizations. It can be so damaging to the patients.

For too long, the global approaches to improve quality and safety gave too little results despite considerable resources.(3) Good intent cannot protect from flawed decisions, poor results and disillusion. Cookbook medicine was not a success and medicine is more complex than monitoring the activity of a doorknobs factory.

The patient needs clinicians who possess sufficient knowledge and clinical skills to make and execute evidence based decisions. Administrators seem to looks for armchair doctors who spend more and more time to send them data to justify for budget cuts because they have the impossible task to extract more from less.

Insanity? Insanity: doing the same thing over and over again and expecting different results (Albert Einstein).

Progress needs research.


1 Saitz R. Candidate Performance Measures for screening for, Assessing, and Treating Unhealthy Substance Use in Hospitals: Advocacy or Evidence-Based Practice? Ann Intern Med. 2010;153:40-3

2 Oxman AD, Sackett DL, Chalmers I, Prescott TE. A surrealistic mega- analysis of redisorganization theories. Healthc Q. 2006;9:50-4

3 Agency for Healthcare Research and Quality. National Healthcare Quality Report, 2008. Chapter 3: Patient Safety. (Accessed Nov 2, 2009, at http://www.ahrq.gov/qual/nhqr08/Chap3.htm)

Conflict of Interest:

None declared

To the Editor:
Posted on July 26, 2010
Jerod M. Loeb, PhD
The Joint Commission
Conflict of Interest: None Declared

Dr. Saitz's recent article Candidate Performance Measures for Screening for, Assessing, and Treating Unhealthy Substance Use in Hospitals: Advocacy or Evidence-Based Practice? (1) lends an important perspective on a topic of critical importance. We agree completely that performance measures must be evidence-based in order to be credible for use in an accountability context (2). However, we believe it is important to note that the candidate performance measures that Dr. Saitz discusses in his article represent a subset of a very early iteration of a candidate performance measure set pertinent to alcohol and tobacco use currently being tested in the field by The Joint Commission. References to drug use have been eliminated from the draft measure set in response to feedback obtained during the stakeholder comment period to which Dr. Saitz refers. As noted in the article, in addition to alcohol use, the draft measure set comprises an equal number of performance measures addressing tobacco use and cessation counseling. A determination as to the ultimate disposition and/or composition of the measure set will be made at the close of the pilot testing period, slated for later this summer. The strength of the scientific evidence behind a performance measure is a crucial component of decisions made by The Joint Commission to include or exclude any measure(s) from a final set.

We would like to emphasize that the development of this performance measure set should not be construed as a veiled attempt to influence public policy, although we acknowledge that some might disagree. Rather, well constructed, well specified, and well tested evidence-based performance measures enable health care organizations to identify gaps in evidence-based care and improve health care quality. We do not anticipate that the alcohol measures that are the subject of Dr. Saitz's article will be mandated for use by all Joint Commission accredited hospitals. Presumably hospitals that have identified a need and desire to improve practice in the areas comprising the measures will adopt them for their own purposes.


1. Saitz R. Candidate Performance Measures for screening for, Assessing, and Treating Unhealthy Substance Use in Hospitals: Advocacy or Evidence-Based Practice? Ann Intern Med. 2010;153:40-3

2. Chassin MR, et al., "Accountability measures" using measurement to promote quality improvement New England Journal of Medicine, 2010, published online June 23; 10.1056/NEJMsb1002320. Available online at:http://www.nejm.org/doi/full/10.1056/NEJMsb1002320

Conflict of Interest:

None declared

The Joint Commission: Advocacy or Evidence-Based Practice
Posted on July 29, 2010
Larry M. Gentilello
University of Texas
Conflict of Interest: None Declared

We read with interest the Perspective article, "Candidate Performance Measures for Screening, Assessing, and Treating Unhealthy Substance Use in Hospitals: Advocacy or Evidence-Based Practice," by Dr. Saitz (1). He concluded that there is insufficient evidence to support screening hospitalized patients for substance use problems, or for offering an intervention or referral to patients that agree to counseling.

He takes the position that clinicians should treat the 70 acute and chronic illnesses that are linked to unhealthy substance use, and ignore the underlying etiology. That would be the death of medical common sense. It would be like treating a myocardial infarction patient without screening for hypercholesterolemia, managing a stroke without measuring blood pressure, treating emphysema without screening for tobacco use, or treating diabetes without measuring glucose concentration.

Dr. Saitz finds flaws in every brief intervention trial by insisting on "perfect evidence." His approach would eliminate the majority of current healthcare practices and send medicine back to the days when doctors had little black bags that contained all of the tools of their trade. He sets up an impossible bar that if followed, would guarantee that no hospitalized patient will ever be screened for unhealthy substance use, much less receive help. He suggests the need for randomized trials for every different type of drug, in every type of combination, without or without concurrent alcohol problems, in every type of medical setting, before evidence of effect is accepted.

The reality is, most patients who misuse any substance tend to misuse multiple substances, and they present to a variety of different types of health care settings. Theories of how people change have been proven to be robust across a variety of settings, cultures, age groups, in both genders, and apply to a variety of risky behaviors. There is little need to endlessly repeat studies while neglecting the very real needs of the suffering patients who are before us. Dr. Saitz criticized screening and brief interventions in trauma centers by stating it was, "based on a single site study that had 54% follow-up for the self-report consumption outcome." Actually, the primary outcome of that randomized trial was injury recurrence. It was reduced by 48% at three years, with close to 100% follow-up (2). As stated in the article, "The use of computerized databases enabled us to obtain objective follow-up information on the primary study outcome variable on all study patients except those who moved out of state." Consumption also decreased by 22 drinks at one year, compared to an 8 drink per week increase in controls (p < 0.01). Do we really want a health system where drunk drivers are admitted to a trauma center and no one offers them counseling to decrease their risk of having another crash?

He discounted another randomized trial by stating, "In a study of 126 patients, a brief intervention-associated decrease in driving while intoxicated was not significant. The decrease became significant in multivariable analyses despite the absence of baseline differences between groups." This is a mischaracterization of the appropriate use of statistics.

A study that does not account for unequal distribution of key predictor variables would be remiss and misleading. Imbalances are typical in small studies despite randomization, and do not have to be statistically significant to affect outcome, or to be appropriately included in a regression model. The investigators controlled for blood alcohol concentration, AUDIT score, age, and prior DUI arrests. At three years there was a 60% reduction in DUI arrests, with one DUI prevented for every nine interventions 2 performed (3). Should patients with unhealthy alcohol use be left to the criminal justice system, or should they receive medical help before they are arrested?

Screening and brief interventions target patients with unhealthy use, not those with late stage addiction. Saitz reported his randomized trial that found no brief intervention effect(4). However, 75% of his patients had severe dependence, one-fourth were homeless, three-fourths had clinical depression, and nearly half had PTSD. The most common diagnoses were alcoholic cirrhosis, cardiomyopathy, gastritis, liver damage, pellagra, polyneuropathy, pancreatitis, withdrawal, convulsions, delirium, hallucinations, psychosis or dementia. These patients are not considered appropriate candidates for screening and brief intervention. These patients would not be counted in the Joint Commission's screening and brief intervention measures.

Dr. Saitz suggests that drug use is not even a health risk by citing studies that did not find a decline in pulmonary function in marijuana users, did not find markers of chronic coronary artery disease in cocaine users, or an increased risk of strokes and other consequences, none of which are surprise findings in studies on acute drug effects. He misquoted the literature by stating that studies do not find psychological damage in young adult drug users. His bias is reflected in his frank dismissal of accumulating evidence that medical risks and consequences are significantly higher among those with alcohol and drug use problems. The Joint Commission measures provide a tool for hospitals to track their screening rates, monitor their compliance with established national guidelines, improve their communication with patients with unhealthy substance use, and to monitor their use of interventions that help, rather than harm patients. Delaying implementation as recommended by Dr. Saitz would pose far greater risks to patients.


1. Saitz R. Perspective. Candidate Performance Measures for Screening for, Assessing, and Treating Unhealthy Substance Use in Hospitals: Advocacy or Evidence-Based Practice? Ann Intern Med. 2010;153:40-43.

2. Gentilello LM, Rivara FP, Donovan DM, Jurkovich GJ, Daranciang E, Dunn CW, Villaveces A, Copass M, and Ries RR: "Alcohol Interventions in a Trauma Center as a Means of Reducing the Risk of Injury Recurrence." Annals of Surgery 230:473-483, 1999.

3. Schermer CR, Moyers TB, Miller WR, Bloomfield LA. Trauma center brief interventions for alcohol disorders decrease subsequent driving under the influence arrests. J Trauma. 2006 Jan;60(1):29-34.

4. Saitz R, Palfal TP, Cheng DM, Horton NJ, Freedner N, Dukes K, Kraemer KL, Roberts MS, Guerriero RT, Samet JH. Brief Intervention for Medical Patients with Unhealthy Alcohol Use. Ann Int Med. 2007;146:167-176.

Conflict of Interest:

Dr. Goplerud and I were technical advisors to the Joint Commission in drafting their substance use performance measures.

Screening for Unhealthy Substance Use in healthcare Settings: A Response to Richard Saitz
Posted on July 30, 2010
Bertha K. Madras
Professor of Psychobiology in the Depart of Psychiatry, Harvard Medical School-NEPRC
Conflict of Interest: None Declared

Of people harboring a medical diagnosis (DSM-IV) of abuse/addiction to alcohol, illicit drugs, or prescription drugs, 95% are unidentified and not treated (1); a larger cohort engages in unhealthy substance use and also remains unidentified. Evidence-based screening, brief interventions (SBI) and referral to treatment services identify and assist a full spectrum of users. To partly address lagging implementation of SBI despite endorsement by Federal agencies and medical associations (2,3), the Joint Commission (JC) is field testing performance measures for delivering alcohol SBI and treatment for alcohol/drug addiction to hospitalized patients (4). Dr. Saitz's Perspective in this Journal (5) questions the Joint Commission's evidence in support of its pilot measures and proffers the view that the JC's advocacy is outpacing evidence-based practice. An accurate description of the current JC measures, a balanced review of the research evidence supporting these measures, and an audit of the statements in the Perspective article, supports different conclusions from those propounded by Dr. Saitz (5):

1. "Measures should be separated for alcohol and drugs because SBI tools and practices differ." Saitz's objections are not current, as he erroneously assumes SBI will be conducted for alcohol and drugs. The JC measures are limited to screening and brief interventions for alcohol and for treatment management and status assessment of alcohol and drug use at, and after discharge (4).

2. "Several assumptions have been made" are erroneous claims: (a) the Joint Commission test measures differentiate between SBI for unhealthy alcohol use (measures 5,6), treatment for substance use disorders (measures 7,8) and rely on published effectiveness research for alcohol (6,7); (b) SBI for drugs is not a JC measure, yet the core features of screening (frequency, amount, consequences) and brief interventions for smoking, alcohol use and illicit drug use (ask, advise, assess, assist, arrange) are the same. (c) For hospitalized patients screened for unhealthy alcohol use, brief counseling is based on evidence from hospitals/emergency departments (e.g. 8-12).

3. "The U.S. Preventive Services Task Force (USPSTF)have shown modest decreases in alcohol consumption at 1 year." His parsimonious view of SBI-associated alcohol decreases contrasts with statistically and clinically significant alcohol use reductions/clinical improvements in intervention groups, as documented and recommended as a prevention service by the United States Preventive Services Task Force (6,7).

4. "The proposed Joint Commission measures cite this evidence (13) as supporting alcohol and drug SBI in inpatients..." The Joint Commission's SBI test measures do not include SBI for drugs. In critiquing the federal effectiveness evaluation, Dr. Saitz omits: (a) data were derived from bundled alcohol/drug screening of 459,599 patients in six states; (b) of these, 104,000 people (22.7%) screened positive for unhealthy substance use and 12,000 patients were randomly selected for six month follow-up, with significant reductions in self-reported alcohol/drug use, cost savings from reduced hospital use, and increased entry into specialty treatment (8, 13-15). (c) He incorrectly states that effectiveness data were based on follow-ups which "lost" 4%-75% of targeted follow-up patients. Data were specifically reported for four states with follow-up rates ranging from 72.3% - 95.5% (13). (d) In the Harris County Hospital District, 59,760 patients admitted over a 39-month period were screened for alcohol/drug use at medical admission, 15,241 (26%) screened positive and received further assessment and services. At a 6-month follow-up interview, drug use and heavy alcohol use decreased substantially from admission to follow-up (8). Effectiveness research is a crucial component of a unified question: does SBI benefit recipients in health-care settings? To reject effectiveness research, which broadens the scope of outcome measures, relaxes inclusion/exclusion criteria, and assesses feasibility of widespread implementation, is to reject a critical component of evaluation.

5. "The relatively low prevalence of drug use is probably another reason." Illicit drug use and prescription medication misuse is common, with past month combined drug use among 16-17 year olds: 15.2%; 18-20 year olds: 21.5%; 21-25 year olds: 18.4%; 26-29 year olds: 13%; 30-34 year olds: 9.6%; 35-29 year olds (1). People between the ages of 18-44 represent 25.14% of hospitalized patients and 40.28% of all ED visits, the latter being highest percentage of all age categories (16).

6. (a) "good reason for concern that brief intervention after drug screening will not be efficacious" (b) plausible that persons identified by screening with occasional marijuana use will have a different response to brief intervention than persons who inject heroin many times daily." No data is cited in support of (a) concern or to sustain (b) assumption.

7. "the proportion of persons identified by drug screening who have dependence is higher than those with unhealthy alcohol use..." The prevalence for alcohol abuse/dependence is 15.2 million, and 7 million for illicit drugs alone or combined with alcohol (1). Among users, the prevalence of illicit drug abuse/dependence (e.g. marijuana, cocaine, 9-10%, 15%) flanks that of alcohol (17,18). The JC measures do not proposed BIs for drugs and BIs are not proposed to treat dependence

8. "The first step toward improving quality of care should be taking care of these patients (with addictions." Although he apparently endorses asking, he proposes that "asking should not be confused with SBI, the use of a validated questionnaire and counseling aimed at substance use."

SBI is a prevention counterpoint to a "don't screen/don't advise" approach. The views articulated by Dr. Saitz are cautious reminders of challenges in segregating science from advocacy/opinion. The Joint Commission's alcohol and drug measures address an important public health problem and measure the delivery of evidence- based practices based primarily on efficacy (RCT) research and supported by effectiveness (naturalistic) research that demonstrate significant health effects. The science supports furthering the measures proposed by the Joint Commission.


1. NSDUH: http://www.oas.samhsa.gov/nsduh/2k8nsduh/2k8Results.pdf

2. Rowland N, Maynard A, Kennedy P, Stone W, Wintersgill W. Doctors and alcohol screening--the gap between attitudes and action. Health Educ J. 1988;47(4):133-6.

3. Kuehn BM. Despite benefit, physicians slow to offer brief advice on harmful alcohol use. JAMA. 2008 Feb 20;299(7):751-3.

4. http://www.jointcommission.org/PerformanceMeasurement/PerformanceMeasurement/Screening and Treating Tobacco and Alcohol Use.htm accessed July 27, 2010.

5. Saitz R. Perspective. Candidate Performance Measures for Screening for, Assessing, and Treating Unhealthy Substance Use in Hospitals: Advocacy or Evidence-Based Practice? Ann Intern Med. 2010; 153:40-43.

6. Whitlock EP, Polen MR, Green CA, Orleans T, Klein J; U.S. Preventive Services Task Force. Behavioral counseling interventions in primary care to reduce risky/harmful alcohol use by adults: a summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med. 2004 Apr 6;140(7):557-68.

7. Solberg LI, Maciosek MV, Edwards NM. Primary care intervention to reduce alcohol misuse ranking its health impact and cost effectiveness. Am J Prev Med. 2008 Feb;34(2):143-152. PubMed PMID: 18201645.

8. The InSight Project Research Group. SBIRT outcomes in Houston: final report on InSight, a hospital district-based program for patients at risk for alcohol or drug use problems. InSight Project Research Group.Alcohol Clin Exp Res. 2009 Aug;33(8):1374-81

9. Havard A, Shakeshaft A, Sanson-Fisher R. Systematic review and meta-analyses of strategies targeting alcohol problems in emergency departments: interventions reduce alcohol-related injuries. Addiction. 2008 Mar;103(3):368-76.

10. Nilsen P, Baird J, Mello MJ, Nirenberg T, Woolard R, Bendtsen P, Longabaugh R. A systematic review of emergency care brief alcohol interventions for injury patients. J Subst Abuse Treat. 2008 Sep;35(2):184 -201

11. Academic ED SBIRT Research Collaborative. The impact of screening, brief intervention, and referral for treatment on emergency department patients' alcohol use. Ann Emerg Med. 2007 Dec;50(6):699-710, 710.e1-6.

12. Bazargan-Hejazi S, Bing E, Bazargan M, Der-Martirosian C, Hardin E, Bernstein J, Bernstein E. Evaluation of a brief intervention in an inner-city emergency department. Ann Emerg Med. 2005 Jul;46(1):67-76.

13. Madras BK, Compton WM, Avula D, Stegbauer T, Stein JB, Clark HW. Screening, brief interventions, referral to treatment (SBIRT) for illicit drug and alcohol use at multiple healthcare sites: comparison at intake and 6 months later. Drug Alcohol Depend. 2009 Jan 1;99(1-3):280-95.

14. Krupski A, Sears JM, Joesch JM, Estee S, He L, Dunn C, Huber A, Roy-Byrne P, Ries R. Impact of brief interventions and brief treatment on admissions to chemical dependency treatment. Drug Alcohol Depend. 2010 Jul 1;110(1-2):126-36. Epub 2010 Mar 26. PubMed PMID: 20347234.

15. Estee S, Wickizer T, He L, Shah MF, Mancuso D. Evaluation of the Washington state screening, brief intervention, and referral to treatment project: cost outcomes for Medicaid patients screened in hospital emergency departments. Med Care. 2010 Jan;48(1):18-24. PubMed PMID: 19927016.

16. AHRQ: http://hcupnet.ahrq.gov/HCUPnet.jsp?Id=EDDF097BB086904A&Form=SelCROSSTAB&JS=Y&Action=%3E%3ENext%3E%3E&_Oneway=No; http://hcupnet.ahrq.gov/HCUPnet.jsp accessed July 28, 2010, 4:23 pm.

17. Warner LA, Kessler RC, Hughes M, Anthony JC, Nelson CB. Prevalence and correlates of drug use and dependence in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry. 1995 Mar;52(3):219-29.

18. Wagner FA, Anthony JC. From first drug use to drug dependence; developmental periods of risk for dependence upon marijuana, cocaine, and alcohol. Neuropsychopharmacology. 2002 Apr;26(4):479-88. PubMed PMID: 11927172.

Conflict of Interest:

None declared

Advocacy versus Evidence-based practice, again
Posted on August 8, 2010
Richard Saitz
Boston University & Boston Medical Center
Conflict of Interest: None Declared

The reader responses to my article could not make any clearer the differences between evidence-based practice and advocacy. The science applicable to the draft performance measures speaks for itself. I am confident the distinctions between evidence and advocacy will be clear to readers of my original article, this letter and the reader responses.

The letter from representatives of the Joint Commission is reassuring. They state that only measures that are supported by high quality evidence will be retained after their testing period. My article points out the lack of an evidence base for drug screening and brief intervention (SBI) performance measures. Drs. Madras, Gentilello and Goplerud argue that the evidence base is sufficient, and criticize me for erroneously evaluating measures of drug SBI when they are not part of the measures. But they were in the publicly posted measures. We now see that after my article went into production a set of revised measures marked as a draft not for distribution appeared, and are referenced by Dr. Madras.

In her letter Dr. Madras provides detail on the prevalence of drug use and on her observational study but no new hospital drug SBI randomized trials which would constitute evidence relevant to a performance measure of that practice. She questions my assertions that drug SBI might be more complex or challenging than alcohol SBI. Clinicians are well aware that identifying and managing prescription drug abuse and sorting out the range of drugs and severities identified is much more difficult than drug SBI. Such clinicians would not be surprised if drug SBI did not work or did not work as well as it does for alcohol, particularly since effect sizes for alcohol SBI are known to be small. The National Institute on Drug Abuse, the Center for Substance Abuse Treatment, and IRBs apparently agree that benefits and risks of drug SBI remain open questions since studies they support and approved are underway to answer them. Advocacy beyond evidence appears to be the explanation for arguments to the contrary. More importantly, the removal of drug SBI from the recent version of the Joint Commission performance measures reinforces the fact that the existing evidence is insufficient for a performance measure.

However, the revised measures continue to include alcohol SBI in hospitals, a practice that is not evidence-based. Before-after studies, even large ones like the one done by Dr. Madras, while informative, should not substitute for randomized trials, particularly when randomized trials are available to inform the questions of interest. Drs. Gentilello and Goplerud imply that my standards for evidence required to support performance measurement are too high and impossible to attain. Space does not permit me to describe or cite the numerous organizations, e.g. US Preventive Services Task Force, National Committee on Quality Assurance, and the Joint Commission, who endorse the need for high levels of evidence before adopting performance measures for preventive services. Published studies addressing in populations for whom there are reasonable concerns about SBI efficacy prove that doing so is not impossible. Drs. Gentilello and Goplerud imply that the evidence is sufficient and that we do not need trials to show hospital alcohol SBI works because we know it works in other settings. This is precisely the misconception and mistaken assumption I emphasize in my article. There are well known clinical reasons like severity and motivational context that suggest the efficacy of SBI will vary by setting. Negative trials in hospitals support those hypotheses.

A systematic Cochrane review finds the evidence for alcohol SBI in hospitals to be inconclusive. Drs. Gentilello and Goplerud discount one study cited in that review, my own, because it did not exclude people with severe dependence, psychiatric conditions or medical illnesses. They say the performance measures would not apply to such patients. But the posted measures note no exclusions for such conditions, nor would they be practical. Excluding such patients would exclude the majority of those identified in hospitals by alcohol screening. More importantly, clinicians doing SBI are aware that when one screens for unhealthy alcohol use, which is defined as the spectrum from risky use through dependence, inclusive, one identifies people who have dependence and other conditions. In fact, in hospitals, the majority of those identified by screening have dependence. Although researchers can exclude such patients from studies, and while the letter writers assert that BI is only for people without such conditions or severity, this is not possible in clinical practice. But even if it were possible, another recent high quality study of hospital alcohol SBI not in the Cochrane review did exclude those with dependence and found no benefit of BI on drinking (1).

The letter from Drs. Goplerud and Gentilello contains much that seriously detracts from an objective discussion of scientific evidence. In my paper I clearly cite the primary outcome of the study by Gentilello et al. Drs. Gentilello and Goplerud in their letter restate these results with two important differences. They neglect to mention that the effect on re-injury was not statistically significant (hazard ratio 0.21-1.29), and that for the secondary drinking outcome, almost half of the sample was lost. They speak of these results as if they were positive evidence of brief intervention efficacy. This type of reporting was recently decried by Boutron et al who find it important to present the negative results of trials clearly (2). In addition, they state that I have a bias and have said that drug use is not a health risk and they go on to criticize references about cocaine and marijuana that they say I have cited. The inaccuracy is so great that one wonders whether they are reacting to another paper. None of this appears in my article raising serious questions about any statements purported to be factual in their letter. I understand that the letter writers may be invested in the measures they contributed to creating but I have no bias that interferes with my summary of the evidence.

They also predict that my approach would be the death of medical common sense, would send medicine back to the old days, and would neglect suffering. I do not agree with these predictions, nor do I think they have much relevance to an objective dispassionate discussion of the evidence that does or does not support a performance measure. Since Drs. Gentilello and Goplerud give their opinion on what I would recommend clinically, and Dr. Madras accurately cites that I would ask about substance use, I believe I should make clear what I do and recommend. I ask my patients if they use alcohol and drugs in all settings for a variety of reasons, such as avoidance of inappropriate prescribing or drug reactions. I counsel and refer patients too. This should not be confused with universal SBI, an apparently common misconception. I implement universal SBI for unhealthy alcohol use in primary care consistent with randomized trial evidence and practice guidelines. I would have us provide the best care to our patients, and I would favor performance measures that support such care, to prevent and reduce alcohol and drug related morbidity and mortality. Although I still have my black bag, I use the latest evidence-based approaches to do so. My record as a clinician and in the peer-reviewed medical literature speaks for itself on this point. In fact, my publicly available academic record supports not only my work as a researcher trying find solutions for such patients, but it also recognizes my work as an advocate for education of clinicians on substance-related conditions and for better patient care. My advocacy proceeds on the basis of evidence, not instead of it or without attention to it.

In developing measures and standards of practice, we should be clear about whether we are proponents of evidence-based practice, or whether we are proposing practices based on things like common sense and strong belief. The latter may have their place, but we should not say the evidence is there when it is not. We owe it to our patients and society to question advocacy and be sure it is supported by evidence. The evidence for hospital performance measures for alcohol (or drug) SBI is either there and largely negative, or simply not there.


1. Freyer-Adam J, Coder B, Baumeister SE, et al. Brief alcohol intervention for general hospital inpatients: a randomized controlled trial. Drug Alcohol Depend 2008:93(3):233.

2. Boutron I, Dutton S, Ravaud P, et al. Reporting and interpretation of randomized controlled trials with statistically nonsignificant results for primary outcomes. JAMA 2010;303:2058.

Conflict of Interest:

None declared

Author's Response
Posted on September 28, 2010
Richard Saitz
Boston University, Section of General Internal Medicine
Conflict of Interest: None Declared

My article (1) says the evidence for efficacy of hospital screening and brief intervention (SBI) is insufficient to support performance measures. The letters from readers confirm this.

It is reassuring the Joint Commission (JC) agrees measures must be evidence-based, and that they deleted drug SBI from theirs (after my article was written)(2). Scientific peer reviewers in federal agencies funding ongoing trials also recognize the evidence lack (3). But the JC measures still contain hospital alcohol SBI, for which evidence is inconclusive (4).

Dr. Madras cites population data which are not relevant to the fact that drug use is less prevalent than unhealthy alcohol use in health settings. She cites her retrospective uncontrolled report which does not inform questions about SBI efficacy. And she objects to characterizing SBI effects as modest. Alcohol SBI is associated with a 10-19% increase in lower risk drinking (5). Readers can decide whether that is modest.

Dr. Madras recognizes SBI is a preventive service, but then says its objective is to identify and motivate people with dependence. Evidence supports the former, but suggests SBI does not improve linkage to treatment for those with dependence (1).

Drs. Goplerud (Co-Chair of the JC Technical Advisory Panel) and Gentilello say there is enough evidence but then say we don't need evidence because patients are suffering and "treatment works" (an inappropriate simplification). Their approach is inconsistent with that stated by the JC, the US Preventive Services Task Force, and other respected groups that require evidence to support guidelines and performance measures.

Drs. Gentilello and Goplerud attribute statements to me that I have not written (e.g. we should ignore etiology, drug use is not risky, it is impossible to do appropriate studies of SBI). I invite readers to read what I actually wrote (1). Their letter is rife with hyperbole, the language of advocacy: "death of medical common sense," "guarantee that no...patient will...receive any help," frequent use of "ever" and "every," "endlessly repeat studies." They make an illogical conclusion--that because drunk drivers may have another crash, we should implement a procedure not known to have efficacy. They presume to know what I would do clinically (which is irrelevant). They confuse performance measurement and clinical practice, incorrectly concluding that a discussion about insufficient evidence for a measure would be the same as discussing clinical care for a patient with a substance-related condition.

They neglect to mention that the primary outcome of Dr. Gentilello's study (6) was not statistically significant, a reporting practice known as "spin" (7). They neglect to mention other negative SBI trauma center trials. An objective summary of the evidence (4 trials with no differences in primary analyses (1)) is not that SBI "works" in trauma centers.

They suggest that details such as setting don't matter. But it matters to patients if we ignore high quality clinical trials that are inconsistent and often negative in hospitals, emergency departments and trauma centers (1, 3). Rather than ignore them, we should learn from them, and do studies to learn when, where and for whom SBI works, and about what we can do in circumstances in which it doesn't work.

Gentilello and Goplerud say hospital SBI will help millions. But when efficacy is unknown, appropriate action is not a performance measure; it is to do appropriate research and to implement what has proven efficacy. Landefeld et al. wrote, "when implemented prematurely [before it is clear that benefits outweigh harms], wishful thinking can replace careful evaluation, and an unproved innovation can become an enduring but possibly harmful standard of care" (8). I am a wishful thinker. But I also know the difference between wishful thinking and evidence.


1. Saitz R. Candidate performance measures for screening for, assessing, and treating unhealthy substance use in hospitals: Advocacy or evidence-based practice? Ann Intern Med 2010; 153:40-43.

2. Draft measures. Screening and treating tobacco and alcohol use. http://www.jointcommission.org/NR/rdonlyres/DE94B4E3-492D-4674-AF36- D618F83F90CD/0/TAM__Tobacco_and_Alcohol_Measure_Set_List.pdf accessed September 23, 2010.

3. Saitz R, Alford DP, Bernstein J, Cheng DM, Samet J, Palfai T. Screening and brief intervention for unhealthy drug use in primary care settings: Randomized clinical trials are needed. J Addict Med 2010;4:123- 130.

4. McQueen J, Howe TE, Allan L, Mains D. Brief interventions for heavy alcohol users admitted to general hospital wards. Cochrane Database Syst Rev. 2009:CD005191.

5. Whitlock EP, Polen MR, Green CA, Orleans T, Klein J; U.S. Preventive Services Task Force. Behavioral counseling interventions in primary care to reduce risky/harmful alcohol use by adults: a summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med. 2004;140:557-68.

6. Gentilello LM, Rivara FP, Donovan DM, Jurkovich GJ, Daranciang E, Dunn CW, et al. Alcohol interventions in a trauma center as a means of reducing the risk of injury recurrence. Ann Surg. 1999;230:473-80.

7. Boutron I, Dutton S, Ravaud P, et al. Reporting and interpretation of randomized controlled trials with statistically nonsignificant results for primary outcomes. JAMA 2010;303:2058-2064.

8. Landefeld CS, Shojania KG, Auerbach AD. Should we use large scale healthcare interventions without clear evidence that benefits outweigh costs and harms? No. BMJ 2008; 336 : 1277

Conflict of Interest:

None declared

Submit a Comment/Letter

Summary for Patients

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.


Buy Now for $32.00

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Related Articles
Related Point of Care
Topic Collections
Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.