0
Ethics in Cardiopulmonary Medicine |

Do Surrogate Decision Makers Provide Accurate Consent for Intensive Care Research?* FREE TO VIEW

Michael Coppolino, MD; Lynn Ackerson, PhD
Author and Funding Information

*From the Department of Medicine (Dr. Coppolino), Kaiser Medical Center, San Francisco, CA; and Kaiser Foundation Division of Research (Dr. Ackerson), Oakland, CA.

Correspondence to: Michael Coppolino, MD, Kaiser Medical Center, Rm 4229, CVICU, 2425 Geary Blvd, San Francisco, CA 94115; e-mail: Michael.Coppolino@ncal.kaiperm.org



Chest. 2001;119(2):603-612. doi:10.1378/chest.119.2.603
Text Size: A A A
Published online

Context: ICU patients are often rendered incapable of making decisions as a result of their illness. The accuracy with which patients’ surrogates consent to research on their behalf is not known.

Objective: To determine if surrogate decision makers provide accurate consent for intensive care research.

Design: Cross-sectional, paired, face-to-face interviews.

Setting: A large, managed-care, cardiac surgery service.

Patients and participants: One hundred elective cardiac surgery patients and their self-appointed surrogates were enrolled.

Intervention: Patients agreed or declined to provide informed consent to two hypothetical research trials. One trial represented minimal risk to those enrolled; the other trial represented greater-than-minimal risk. Surrogates attempted to predict the patients’ responses.

Main outcome measures: The accuracy of surrogate consent was analyzed in a fashion analogous to the evaluation of a diagnostic test. Predictors of accuracy were evaluated using multiple logistic regression.

Results: Overall surrogate positive predictive value for the low-risk study was 84.0% and for the high-risk study was 79.7% (p = 0.72, McNemar test). Predictors of accurate consent were not consistent across the two studies.

Conclusions: Surrogate decision makers for critical-care research resulted in false-positive consent rates of 16 to 20.3%. Further assessment and evaluation of the practice of surrogate consent for intensive care research is, therefore, recommended.

Medical research involving individuals with diminished decision-making capacity is a topic of international controversy.12 In the United States, the absence of universally applicable national standards regarding research involving patients unable to provide their own consent has resulted in local supervision of such research. This approach has been problematic, as evidenced by a series of reports by the Inspector General of the Department of Health and Human Services, concluding that the research supervision process was in need of reform.3

Although the problems with this delegation of federal authority have been evidenced largely in the field of psychiatric research, critically ill patients who have become incapable of making decisions as a result of the severity of their illness present analogous problems.45 Policies enabling ethical consent to research are needed if critically ill patients are to be protected as vulnerable subjects and yet not be excluded from realizing the potential benefits of research.

Several strategies have been identified for the ICU setting in which the patient cannot provide voluntary informed consent for himself or herself: (1) consent could be deferred and obtained retrospectively if and when the patient regains capacity, (2) consent could be waived entirely under the auspices of beneficence, (3) consent could be obtained from a surrogate, (4) consent could be obtained prospectively (analogous to an advance directive for therapeutic decisions).6 The practice of obtaining surrogate consent for research on critical-care patients incapable of making decisions has become the pragmatic standard in the United States.

The objectives of the present study were as follows: (1) to determine if surrogate decision makers provide accurate consent (that which agrees with the patient’s wishes) for critical-care research trials involving minimal risk and greater-than-minimal risk, and (2) to identify potential predictors of accuracy in the surrogate decision-making process.

Participants

All patients who were scheduled for elective cardiac surgery at the Kaiser San Francisco Medical Center, San Francisco, CA, and their respective surrogates were considered for participation in the study. This medical center serves as the referral center for all cardiovascular surgery performed on Kaiser Health Plan members residing in the Northern California service area. Cardiologists refer patients for cardiac surgery using a centralized booking office at the San Francisco center. This central office was the sole source of recruitment. Patients who were emergently scheduled from the cardiac catheterization laboratory or who urgently transferred from other hospitals were not considered for enrollment.

A single interviewer contacted potential participants to arrange interviews. All patients aged ≥ 18 years at the time of the interview were eligible if they were scheduled for any cardiac surgery procedure. Patients were excluded if they did not provide informed consent or if they were unable to communicate in English, were unable to identify a surrogate decision maker, or were unavailable for interview in person. To establish a baseline level of cognitive function, all patients and surrogates were required to correctly answer 12 questions on a 14-question cognitive screen as modified by the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments investigators.7

Patients were asked to identify a surrogate who they would choose to make medical treatment decisions. Patients were informed that preferred surrogates were (given in descending order) an individual who has obtained written power of attorney for health care; a spouse or partner (a partner is defined as someone with whom the patient has lived for at least 1 year and has a close personal relationship that is of primary importance in both persons’ lives); the patient’s child; the patient’s parent; the patient’s brother or sister; any other relative of the patient; and friend of the patient. Surrogates were eligible to participate unless they met any of the exclusion criteria described for patient enrollment.

Measurements

Data were gathered by face-to-face interviews with patients and surrogates. A sole interviewer performed all data collection sessions. Patients and surrogates were separated for review of study materials and concurrent completion of data entry sheets and remained blinded to the others responses. Each participant completed a cognitive screen7 and a demographic survey. Surrogates were asked whether or not they had ever had discussions with the patient involving the patients’ views regarding artificial life support or the patients’ preferences to be involved in medical research.

Participants reviewed two study protocols and respective consent forms that were modified versions of previously completed critical-care research trials (Appendix). After reviewing the study protocols, patients were reminded that the studies were hypothetical and they would not actually be entered in the studies during their impending hospitalization. Patients were asked, “Having reviewed the study descriptor and consent form, would you agree to participate in this study if you became eligible during your intensive care stay?” Responses for both studies were recorded on a 4-level Likert scale (definitely yes, probably yes, probably no, definitely no). Surrogates were asked, “Based on what you believe would be the wishes of the patient to whom you are surrogate (ie, not necessarily your own views) and having read the study descriptor and consent form, would you wish to consent to enroll your patient to participate in this study if they became eligible during their intensive care stay?” Surrogates responded using the same 4-level Likert scale.

Surrogates also responded to the question “How certain are you that your response regarding research participation accurately reflects the wishes of the patient for whom you are the surrogate?” using a similar Likert scale (very certain, certain, uncertain, absolutely uncertain).

The Anti-A study was believed to require minimal deviation from standard care and represent minimal risk (subjects receive one of two drugs with very similar efficacy and few side effects), and the KAM 200 study represented greater-than-minimal risk (subjects receive active drug or placebo for extreme illness, including failure of vital organs; potential drug side effects include damage to the liver and unknown risks to the body).8 Both studies presented the potential for direct benefit to the research subject.

Before patients and surrogates were dismissed, all data-entry forms were reviewed by the interviewer to ensure completeness and legibility. Each patient/surrogate pair received a $25 stipend to compensate for extra time spent at the medical center and parking expenses.

Statistical Analysis

The primary end point, the accuracy with which surrogates predicted the preferences of patients to enroll in ICU research trials, was assessed using a strategy analogous to that used for the evaluation of diagnostic tests.9 A patient’s decision to enroll in a study represents the “truth” or the “gold standard.” The surrogate’s prediction of their respective patient’s enrollment preferences represents the test result. Because in the critical-care arena the consent decision (the “gold standard”) of the patient who is incapable of making decisions would not be known, the positive and negative predictive values were calculated.

When a surrogate consents to enroll their respective patient into a study, the test result is considered positive. When a surrogate declines to provide consent the test result is considered negative.

The positive predictive value represents the proportion of patients with a positive test result who have the target condition. In this study, the positive predictive value is the proportion of surrogates who agree to enroll their patients in the research trial (positive test) when the patient also would have agreed to enroll (target condition). For societal reasons,4 we were particularly interested to know how frequently surrogates would consent to enroll patients in research when the patient would not have done so had they been able to provide their own informed consent (known as the false-positive rate of the test).

To generate a dichotomous outcome of agreement between patients and surrogates, the Likert responses of “definitely yes and probably yes” were collapsed to “yes,” and “probably no and definitely no” were collapsed to “no.”

Sample size was based on an expected overall enrollment of 75% of patients. If the positive predictive value of the surrogates was 70%, a sample of 100 pairs would allow us to estimate this positive predictive value within 10%.

In the secondary analyses, the characteristics of those surrogates who accurately predicted consent preferences were compared for both study scenarios. Thresholds for aggregating continuous variables were determined by exploration of the distribution of responses for each item, and a dichotomized analysis was used wherever possible in an attempt to avoid multiple comparisons. Medians and interquartile ranges were used to summarize continuous variables due to the skewed distributions of some of these variables. The Wilcoxon rank sum test was used to compare continuous variables, and theχ 2 test or Fisher’s Exact Test was used for categorical variables. Paired data were examined using McNemar test for paired comparisons. A stepwise multivariate logistic regression model of agreement between surrogates and patients was constructed using those variables that were statistically significant at the 0.15 level in the bivariate analysis.

Multivariate results were summarized as odds ratios (ORs) and 95% confidence intervals (CIs). Significant levels were set atα = 0.05, and no adjustments were made for multiple comparisons.10 Statistical software packages (SAS PC; SAS Institute; Cary, NC and SPSS 8.0; SPSS; Chicago, IL) were used for analyses.

Research Ethics

This study was approved by the Kaiser Foundation Research Institute institutional review board. Written informed consent was obtained from all patients and surrogates.

Accrual of Patients and Patient Characteristics

Between February 24, 1998, and November 13, 1998, 530 patients had elective cardiac surgery arranged through the cardiac surgery office at the Kaiser San Francisco Medical Center. During this period, 320 patients (60%) were screened for eligibility. Patients were not screened when the interviewer was unavailable. There were 290 eligible patients. Reasons for ineligibility included the following: unable to communicate in English (n = 19), no available surrogate (n = 7), severe visual impairment (n = 3), and other (n = 1). Of the eligible patients, 74 presented without an eligible surrogate, 66 patients and/or surrogates refused participation, 36 patients were not entered due to administrative/logistic problems, and 14 patients were required to attend other appointments. Eligible patients not enrolled in the study were demographically similar (median age, 64.0 years; coronary artery bypass graft [CABG] surgery, 56.5%; valve surgery, 24.1%; combined valve-CABG surgery, 16.8%; and other, 2.5%) to those enrolled. All patients and surrogates passed the cognitive screening test.

Table 1 shows patient and surrogate baseline characteristics. Surrogates were younger and more likely to be female than were patients but had similar educational levels, race/ethnicity, and religious affiliations.

The median patient age was 61.5 years, while the median for surrogates was 54.5 years. Sixty-eight patients (68%) and 66 surrogates (66%) volunteered their race/ethnicity. Patients were predominantly white (70.6%) and male (73%), and surrogates were predominantly white (60.6%) and female (75%).

The patient/surrogate relationship was reported as spouse, 47%; child, 23%; power of attorney for health care, 17%; friend, 7%; and brother, sister, or other relative, 4%. The median duration of these relationships was 36 years. Seventy percent of surrogates reported having had previously discussed life-support preferences. Thirty percent recalled having previously discussed participation in medical research.

Sixty-three percent of patients attended college, and 37% achieved grade 12 or less. For surrogates, these educational levels were 61% and 39%, respectively. Table 2 shows the patients’ medical demographics. Fifty-two percent of patients had CABG surgery, 29% had valve surgery, 14% had combined valve-CABG surgeries, and 4% had other cardiac surgeries. The most common indication for surgery was unstable angina. Common premorbid conditions included hypertension (58%), previous myocardial infarction (23%), diabetes mellitus (22%), and congestive heart failure (21%).

Outcome Events (Substitute Decision Making)
Preferences:

Overall hypothetical patient and surrogate enrollment in the two research protocols is presented in Table 3 . There was no difference in the proportion of patients enrolled in the Anti-A study (65%) and the KAM 200 study (74%; p = 0.82, McNemar test).

Table 4 shows the accuracy with which surrogates predicted the preferences of patients to enroll in the two critical-care research trials. In the Anti-A study, with the surrogates’ substituted judgment considered as a diagnostic test for the “gold standard” patients’ decision, the positive and negative predictive value of surrogates’ responses were 84.0% and 51.4%, respectively. For the KAM 200 study, the positive and negative predictive values for the surrogates’ responses were 79.7% and 46.2%, respectively. The sensitivity of the surrogates’ responses for the Anti-A study was 76.4% with a specificity of 64.0%. Sensitivity and specificity for the KAM 200 study were 80.4% and 44.4%, respectively.

Using the 4-level response scale for the Anti-A study, surrogates correctly predicted the patients’ responses 52% of the time, were off by one category 36% of the time, by two categories 11% of the time, and by three categories in 1% of cases. The accuracy of the respective predictions for the KAM 200 study was 45%, 42%, 10%, and 3%.

The predictive accuracy of the surrogates did not differ between the low-risk Anti-A study and the high-risk KAM 200 study (p = 0.72, McNemar test).

The bivariate analysis for the predefined patient and surrogate subgroups is illustrated in Table 5 . For the Anti-A study, the only subgroup in which surrogates showed a significantly higher positive predictive value was if the patients’ age was greater than the median patient age (61 years). For the KAM 200 study, a significantly higher positive predictive value was associated with the surrogate relationship being durable power of attorney (DPA) for health care, surrogates recalling previous discussions regarding life support, and surrogates feeling certain about their predictions.

Gender, duration of relationship, discussions about research involvement, education, race/ethnicity, religious affiliation, and the order in which the study scenarios were presented to the respondents were not significant predictors of surrogate accuracy in the bivariate analysis.

Independent predictor variables determined in the bivariate analysis (Table 5) were used to construct a multivariable model to determine which combination of variables predicted patient/surrogate agreement to be involved in ICU research.

The statistically significant variables entered into the model for the Anti-A study were patient age greater than median age, patient/surrogate age difference within 5 years, and patient educational level less than college. For the KAM 200 study, the variables entered were patient age greater than the median age, surrogate certainty about their decision, surrogate recall of a discussion with the patient about life-support issues, patient/surrogate relationship, surrogate race/ethnicity, and patient race (white).

The results of the multivariate model are presented in Table 6 . For the Anti-A study, there was less agreement between surrogates and patients if the age difference between patients and surrogates were> 5 years (OR, 0.4; 95% CI, 0.1 to 0.8; p = 0.02) and there was improved agreement if the patient did not attend college (OR, 3.2; 95% CI, 1.1 to 10.6; p = 0.04).

For the KAM 200 study, agreement between patients and surrogates was higher for surrogates who reported discussing life-support issues (OR, 4.1; 95% CI, 1.6 to 11.2; p = 0.004) and for surrogates who felt“ absolutely certain” or “certain” about their predictions (OR, 3.6; 95% CI, 1.4 to 9.8; p = 0.01).

Intensive care research trials frequently use surrogate consent to enroll critically ill patients who lack decision-making capacity. Questions exist regarding both the process and the moral basis of this form of consent.4 The documented inaccuracies in surrogate decision making for treatment decisions1115 and the lack of empirical evidence in the research setting motivated the present study.

Our results, demonstrating surrogate inaccuracies in predicting patient preferences for research involvement, highlight the concerns with surrogate consent. The present study utilized face-to-face interviews, allowing the opportunity for questions and explanations; used detailed and explicit consent forms; prompted surrogate responses based not on their personal beliefs but on what they believed would be the wishes of the patient; and required all surrogates to pass a cognitive screening test. Given this methodology, we find it particularly concerning to demonstrate that 16 to 20.3% of surrogates agreed to enroll patients into research trials that the patients would not have entered had they been capable to make their own consent decisions.

There are several potential explanations for these inaccurate predictions. Firstly, for treatment decisions, evidence exists that surrogates who discuss preferences beforehand have improved predictive accuracy.1213,16 It is conceivable that a similar phenomenon may exist for research decisions. Few surrogates (30%) in our study had previously discussed research preferences; therefore, the study had limited power to demonstrate improved predictive accuracy in this subgroup. Secondly, the complex nature of critical illness and the conduct of research trials have led some to question whether consent is ever truly “informed.”4,1719

In the absence of a standard capacity assessment, there has been little empirical evaluation of the level of comprehension necessary for informed decision making for involvement in medical research. We used the cognitive screen as used in the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment7 in an attempt to be consistent with current research practices, but in no way claim that such a competency evaluation demonstrates capacity to understand the consent process for ICU research involvement.

Thirdly, although all efforts were made to ensure that surrogates were acting on behalf of their respective patients in the consent process, there are many forces that may compromise surrogates’ abilities to do so. A critical illness involving their loved one undoubtedly results in a complex emotional experience that may confound surrogates’ representation of patients’ views. Surrogates may enroll patients in an attempt to be in agreement with the medical team at times of serious illness. Such forces would be expected to result in surrogates making false-positive consent decisions.

No other literature exists with which to compare the surrogate accuracy as demonstrated in our study. Examples involving surrogate predictions of patient preferences for treatment decisions in the setting of chronic medical illness have shown predictive accuracy in the 53 to 90% range.1314 Results of these studies involve a heterogeneous collection of therapeutic decisions in widely varied clinical scenarios and cannot be applied to decisions regarding consent to research.

The present study evaluated a low-risk study scenario and a high-risk study scenario, both of which presented the potential for direct benefit to the patient. This study design was an attempt to evaluate the influence of risk on the accuracy of surrogate consent because regulators often recommend that research involving higher levels of risk require a more conservative (eg, legally appointed surrogates) approach to the designation of surrogates.8,20 Our study did not demonstrate differences in the accuracy of surrogate predictions between the high-risk and the low-risk study; therefore, a higher standard for the surrogate consent process for high-risk research protocols can neither be supported nor refuted on the basis of our study. How patients and surrogates perceive the risks of clinical research is not known, and the current investigation did not attempt to evaluate these perceptions.

The secondary objective of our study was to determine clinically identifiable subgroups in which surrogate decision making was more or less accurate. Superior predictive accuracy was expected a priori from three subgroups: surrogates who felt particularly certain about their predictions, surrogates who had previously discussed life support with their respective patients, and surrogates who had previously discussed research preferences with their respective patients. Multivariate analysis demonstrated that surrogates who felt certain of their predictions and who had prior discussions regarding life support showed greater patient-surrogate agreement only for enrollment in the KAM 200 study. Although investigators may feel more comfortable that these patients’ wishes are more accurately being represented, we remain cautious about this conclusion given the lack of improved accuracy in the Anti-A study.

Limitations

We recognize several limitations of our study. The described inaccuracies of surrogate decisions may reflect uninformed decision making or may be the result of methodologic features of our study. All scenario studies are somewhat artificial, and how our results would compare to patients and surrogates preferences in real-time ICU research trials is not known. We felt ethically obligated to inform subjects that they would not actually be entered in the described studies. Also, although surrogates and patients may make different decisions in actual situations, there is no reason to believe that the accuracy of surrogate predictions would necessarily change in these situations. Hypothetical scenarios will remain an integral part of research on advance directives and surrogate decision making because the thoughts of subjects who are incapable of making decisions cannot be directly ascertained.

A second limitation of our study was that the subjects in our study were awaiting elective cardiac surgery at a single referral center with the requisite ICU admission. Caution should be exercised when extrapolating beyond this population. Also, participants unable to communicate in English were not enrolled because of concerns regarding the validity of translations of study material.

A third potential limitation involved the use of a 4-level Likert response format for patient and surrogate responses. Research consent forms generally require a yes or no response; however, we allowed subjects to decide if they would “definitely” or “probably” enter or “definitely” or “probably” not enter the study. Likert responses were used to be consistent with previous research efforts11 and to improve qualitative exploration of our data. Subjects were not provided with an “undecided” response option.

A fourth limitation was our inability to screen all elective surgery patients due to lack of interviewer availability. Eligible but not enrolled patients were demonstrated to be demographically similar to those enrolled. No data were collected on patients not screened for enrollment.

Despite these limitations, our study raises serious concerns for investigators, institutions, and governing bodies because 16 to 20.3% of surrogates would have agreed to enroll patients in intensive care research trials against the patients’ personal preferences had they maintained decisional capability. Investigators need to ensure that the surrogate consent process is structured to minimize inaccuracies and yet not deprive critically ill patients the potential benefits of research. Institutions and governing bodies will need to decide what degree of surrogate inaccuracy is acceptable for continued approval of research projects involving surrogate consent. If such guidelines are dependent on the degree of risk and the potential benefits of the research trial, then this risk-benefit ratio should be determined from the patients’ perspective. The level of accuracy of surrogate decision making demonstrated in our study provides a baseline estimate for intensive care research; however, to maintain public confidence in the process of medical research, these findings should trigger the evaluation of (1) strategies designed to encourage potential patients and surrogates to openly discuss their views regarding medical research,21 (2) the development of advanced directives for medical research participation, (3) the appointment of DPA for involvement in medical research. Research of this nature would provide a useful context in which patients and their surrogates could begin to discuss these important health-care issues.

Description for KAM 220 Study
Introduction:

I have been told by my doctor that I (or my friend or relative) have a severe case of “sepsis syndrome.” It has been explained to me that sepsis syndrome is a complex set of symptoms that occur as a result of life-threatening infection, and that lead to the failure of vital organs, such as the heart, liver, and kidneys. I have been informed that in North America, approximately 100,000 deaths related to sepsis occur annually.

It is generally believed that sepsis syndrome is caused by an overly active immune response, the response that normally clears infection from the body. This exaggerated response may lead to inadvertent injury of body tissue and the eventual failure of the vital organs.

The experimental medication that is being studied is called KAM 200F. KAM 200F neutralizes some of this overly active immune response. I have been asked to participate in this research study to evaluate KAM 200F in the treatment of sepsis syndrome.

Purpose of the Study:

The aim of this study is to see whether KAM 200F, in addition to the standard treatment for severe sepsis, will improve survival for patients with sepsis syndrome. The study will evaluate the safety and effectiveness of KAM 200F when it is used to treat patients with sepsis syndrome. I have been informed that this research study is taking place in various centers across the United States and Canada and that I will be one of approximately 1,500 participating patients to receive the study medication or a placebo (inactive) medication. Both KAM 200F and placebo treatment contain very small amounts of human serum albumin.

Treatment:

As a participant in this study, I will receive either KAM 200F or a placebo drug. I understand that whether or not I receive KAM 200F will be decided by random chance, and neither I nor the staff involved in my care will know which treatment will be given. I will receive an intravenous dose of KAM 200F or placebo every 8 hours, on nine occasions—for a total 72-hour treatment period.

In addition to the study medication, I will receive all appropriate standard therapies for my condition because KAM 200F is intended to supplement, not replace, standard therapy. I will be closely followed by my doctor and other personnel of the 72 hours that I am receiving the treatment, as well as for the 25 days thereafter.

Tests and Procedures:

I am aware that as part of my participation in this study, I will be undergoing a number of tests. Many of these tests are part of the routine care of patients with sepsis syndrome and include physical examination, blood tests, tests of the function of many body organs, chest x-rays, and electrocardiograms. In addition, blood samples will be taken for the special tests related to the research study. The total amount of extra blood will be approximately 208 mL (each 5 mL is approximately one teaspoon) over the entire 28 days of the study. If I am discharged from the hospital before 28 days have passed since receiving KAM 200F, I can expect to be contacted by my doctors and nurses to find out the status of my health at that time. I may also be requested to return to the hospital or doctor’s office to complete the blood tests, which would consist of approximately 5 mL (one teaspoon) of blood to test for antibiotics.

Benefits and Risks:

Participating in this study may reduce the damage to my vital organs resulting from sepsis syndrome and, therefore, may increase my likelihood of survival; however, it is entirely possible that participating in the KAM 200F study may not benefit me in any way.

There may be risks and discomfort associated with my participation in this study. Blood samples required for the study will be drawn from an indwelling catheter, called an arterial line, or from standard needle sticks from the veins in my arms. My body will normally replace the amount of blood taken for these studies in a short period of time.

The experience of the research thus far using KAM 200F in humans has provided no direct evidence that the drug itself causes harmful effects. In the clinical trails to date, approximately 266 patients have received KAM 200F. A minimal elevation of liver enzymes (which could indicate damage to the liver) was found in two patients, but this increase was thought to be due to the patient’s underlying disease. One patient experienced slowing of the heart rate, which was thought to possibly be related to KAM 200F. This condition later improved. One patient experienced a rash over the right groin and hip area, which was thought to be related to KAM 200F. This condition also improved.

Treatment with KAM 200F may result in other unknown risks to my body. All reasonable precautions will be taken to remove or reduce these risks, and I will be closely monitored by my care givers. I have been made aware that there is no information about the safety of this drug in unborn children and that I should not participate in this study if I am, or think I may be, pregnant. If I am a woman of child-bearing age, a pregnancy test will be performed to rule out pregnancy.

There is no way to be certain that I will benefit from this study, although the results of this study may in the future benefit other patients with sepsis syndrome. Because sepsis syndrome is not yet fully understood, there is a possibility that I will either not benefit or may possibly even have a worse outcome than if I do not participate in the study.

Alternative Treatments:

If I choose not to enter this study, my doctor will treat me with standard medications and I will receive the care for sepsis that he or she would normally provide. My participation is voluntary, and I may withdraw from the study at any time for any reason. As well, my physician may withdraw me from the study at any time if he or she feels that it is not in my best interest to remain in the study. This will not affect my ongoing care in any way. In addition, the study sponsor or investigator may discontinue my participation in the study at any time.

Confidentiality of Records:

Although representatives of the study sponsors, the US Food and Drug Administration, and the hospital ethics committee where I am being treated may wish to examine my records, I have been assured that my name will not be revealed in any scientific publication resulting from this study, without my expressed consent. I will be identified by my initials and code numbers only.

I have read the informed consent form and have discussed the study with the consent interviewer, and my questions have been answered to my satisfaction. I may contact Dr. Michael Coppolino (the principal investigator), at xxx-xxx-xxxx, at any time regarding questions about the study.

Description of the Anti-A Study
Purpose of the Study:

I have been told by my doctor that I (or my relative or friend) have a high risk of developing bleeding ulcers in my stomach. Patients admitted to an ICU who require support with a ventilator (breathing machine) for more than 2 days may develop bleeding inside the stomach from stress ulcers in the stomach. The risk of significant bleeding from these ulcers is approximately 5 to 10%, but can be reduced by the use of medications that block production or the effect of the stomach acid. Because patients at risk for bleeding are also at risk for the development of pneumonia, we wish to determine whether the risk of pneumonia is related to the type of medication used to prevent bleeding. Several studies have suggested that those medications that coat the stomach in order to protect it from acid, rather than those medications that protect the stomach by preventing production of acid by the stomach, may reduce the risk of pneumonia.

Although both types of drug are in wide use and are equally effective in preventing bleeding, we do no know which one is better than the other for preventing pneumonia. The purpose of this study, therefore, is to compare Anti-A (a drug that blocks acid production) with Coat-M (a drug that coats the stomach) to see if one or the other will reduce rates of ICU pneumonia in patients who need these medications to reduce stress ulcer bleeding.

Patients identified as facing an increased risk for stress bleeding would normally be given some form of preventive medication. Should you consent to participate in the study, you will be assigned to receive either Anti-A or Coat-M on a random basis. Neither you nor the doctors looking after you, nor the study personnel will know which medication you are receiving so that evaluation of the drugs will not be biased. Coat-M will be given by mouth while Anti-A will be given intravenously. A placebo form of the drug you are receiving will be administered and those medications will be administered for the duration of your ICU stay.

If your doctor or the ICU doctors suspects that you have developed pneumonia, samples of fluid from the lung will be taken (after you are comfortable sedated with medication) by passing a flexible tube called a bronchoscope into the lung. This technique is standard practice for diagnosing pneumonias in the ICU.

Apart from also receiving a placebo drug, your ICU care will not differ from that which would be given to you should you choose not to participate in the study.

Discomforts and Risks:

Although both of the medications are licensed for use in preventing stress ulcer bleeding and have been widely used around the world for many years, like all drugs they do have potential side effects. Side effects with Coat-M are uncommon and consist primarily of stomach upset in the form of constipation, nausea, vomiting, and diarrhea. Patients receiving Anti-H can develop abnormalities of liver function, although these changes are very uncommon. You will be carefully monitored for the development of possible side effects. Should these be detected, the medications will be discontinued.

Bronchoscopy to diagnose pneumonia often causes coughing because of irritation of the breathing tube by the bronchoscope. Occasionally, scant bleeding or blood streaking of phlegm may result following the bronchoscopy.

Benefits:

It is possible that your chance of developing pneumonia will be lessened because of participating in this study; however, it is entirely possible that no direct benefit will come to you as a result of your participation in this study. We do not expect, however, that the results of this study will provide important information in deciding the best medications for patients in the future.

Confidentiality:

You will not be identified by name in any reports arising from this study. Your identity will be known only to the doctor looking after you, the study investigators and, if necessary, a committee of study investigators who are judging outcome. The study sponsor may need to see you records to answer any questions about the study.

Relation to Your Ongoing Care:

Your doctor knows about this study and consents to participation in it. Should your doctor so request, you could be immediately withdrawn from the study. Similarly, if you or a family member wish to withdraw from the study at any time, you may do so without in any way jeopardizing your ongoing care.

Abbreviations: CABG = coronary artery bypass graft; CI = confidence interval; DPA = durable power of attorney; OR = odds ratio

This study was supported by Kaiser Foundation Community Services Grant No. 131-9770, which enabled payment for statistical support (Dr. Ackerson), an interviewer, and the patient-surrogate stipend.

Table Graphic Jump Location
Table 1. Characteristics of Patients and Surrogates*
* 

Data are presented as median (25th, 75th percentiles) or % unless otherwise indicated.

Table Graphic Jump Location
Table 2. Clinical Demographics of Patients
* 

Data are presented as median (25th, 75th percentile); ASD = atrial septal defect; VSD = ventricular septal defect.

Table Graphic Jump Location
Table 3. Overall Enrollment Preferences of Patients and Surrogates in Anti-A and KAM 200 Studies
Table Graphic Jump Location
Table 4. Summary of Surrogates’ Overall Predictive Accuracy*
* 

PPV = positive predictive value; NPV = negative predictive value; A = Anti-A; K = KAM 200. No statistical difference between positive predictive value between Anti-A study and KAM 200 study (McNemar test, p = 0.72).

Table Graphic Jump Location
Table 5. Accuracy of Surrogate Decision Making by Subgroup*
* 

Sens = sensitivity; Spec = specificity; see Table 4 for other abbreviations.

 

undefined (denominator = 0).

Table Graphic Jump Location
Table 6. Logistic Regression Analysis of Significant Predictors of Agreement Between Surrogate and Patient

We thank Nydia Lovell for completing the study interviews and Dr. E. Etchells for his assistance with protocol development.

Shuster, E (1997) Fifty years later: the significance of the Nuremberg code.N Engl J Med337,1436-1440. [CrossRef] [PubMed]
 
Matot, I, Pizov, R, Sprung, CL Evaluation of Institutional Review Board review and informed consent in publications of human research in critical care medicine.Crit Care Med1998;26,1596-1602. [CrossRef] [PubMed]
 
Capron, A Ethical and human-rights issues in research on mental disorders that may affect decision-making capacity.N Engl J Med1999;340,1430-1434. [CrossRef] [PubMed]
 
Silverman, HJ Ethical considerations of ensuring an informed and autonomous consent in research involving critically ill patients.Am J Respir Crit Care Med1996;154,582-586. [PubMed]
 
Basler, JR, Martinez, EA, Winters, BD β-Adrenergic blockade accelerates conversion of postoperative supraventricular tachyarrhythmias.Anesthesiology1998;89,1052-1059. [CrossRef] [PubMed]
 
Pinder, M, Tshukutsoane, S, Scribante, J, et al Critical care research and pre-emptive informed consent: a practical approach used in Chris Hani Baragwanath ICU.Intensive Care Med1998;24,353-357. [CrossRef] [PubMed]
 
Bergner, M Patient characteristics in SUPPORT: activity status and cognitive function.J Clin Epidemiol1990;43,S33-S36. [CrossRef]
 
Brody, BA The ethics of biomedical research: an international perspective.1998,300-302 Oxford University Press. New York, NY:
 
Sackett, DL Clinical epidemiology: a basic science for clinical medicine 2nd ed.1991,69-152 Little, Brown and Company. Boston, MA:
 
Rothman, K No adjustments are needed for multiple comparisons.Epidemiology1990;1,43-46. [CrossRef] [PubMed]
 
Uhlmann, R, Pearlman, R, Cain, K Physicians’ and spouses’ predictions to elderly patients’ resuscitation preferences.J Gerontol1988;43,M115-M121. [PubMed]
 
Sulmasy, DP, Terry, PB, Weisman, CS The accuracy of substituted judgments in patients with terminal diagnoses.Ann Intern Med1998;128,621-629. [PubMed]
 
Ouslander, J, Tymchuk, A, Rahbar, B Health care decisions among elderly long term care residents and their proxies.Arch Intern Med1989;149,1367-1372. [CrossRef] [PubMed]
 
Hare, J, Pratt, C, Nelson, C Agreement between patients and their self-selected surrogates on difficult medical decisions.Arch Intern Med1992;152,1049-1054. [CrossRef] [PubMed]
 
Emanuel, E, Emanuel, L Proxy decision making for incompetent patients.JAMA1992;267,2067-2071. [CrossRef] [PubMed]
 
Sulmasy, DP, Haller, K, Terry, PB More talk, less paper: predicting the accuracy of substituted judgements.Am J Med1994;96,432-438. [CrossRef] [PubMed]
 
Epstein, LC, Lasagna, L Obtaining informed consent - form or substance?Arch Intern Med1969;123,682-688. [CrossRef] [PubMed]
 
Truog, RD, Robinson, W, Randolph, A, et al Is informed consent always necessary for randomized controlled trials?N Engl J Med1999;10,804-807
 
Etchells, E, Sharpe, G, Walsh, P Bioethics for clinicians: 1. Consent.Can Med Assoc J1996;155,177-180
 
Michels, R Are research ethics bad for our mental health?N Engl J Med1999;340,1427-1430. [CrossRef] [PubMed]
 
Gordon, N, Shade, S Advance directives are more likely among seniors asked about end-of-life care preferences.Arch Intern Med1999;159,701-704. [CrossRef] [PubMed]
 

Figures

Tables

Table Graphic Jump Location
Table 1. Characteristics of Patients and Surrogates*
* 

Data are presented as median (25th, 75th percentiles) or % unless otherwise indicated.

Table Graphic Jump Location
Table 2. Clinical Demographics of Patients
* 

Data are presented as median (25th, 75th percentile); ASD = atrial septal defect; VSD = ventricular septal defect.

Table Graphic Jump Location
Table 3. Overall Enrollment Preferences of Patients and Surrogates in Anti-A and KAM 200 Studies
Table Graphic Jump Location
Table 4. Summary of Surrogates’ Overall Predictive Accuracy*
* 

PPV = positive predictive value; NPV = negative predictive value; A = Anti-A; K = KAM 200. No statistical difference between positive predictive value between Anti-A study and KAM 200 study (McNemar test, p = 0.72).

Table Graphic Jump Location
Table 5. Accuracy of Surrogate Decision Making by Subgroup*
* 

Sens = sensitivity; Spec = specificity; see Table 4 for other abbreviations.

 

undefined (denominator = 0).

Table Graphic Jump Location
Table 6. Logistic Regression Analysis of Significant Predictors of Agreement Between Surrogate and Patient

References

Shuster, E (1997) Fifty years later: the significance of the Nuremberg code.N Engl J Med337,1436-1440. [CrossRef] [PubMed]
 
Matot, I, Pizov, R, Sprung, CL Evaluation of Institutional Review Board review and informed consent in publications of human research in critical care medicine.Crit Care Med1998;26,1596-1602. [CrossRef] [PubMed]
 
Capron, A Ethical and human-rights issues in research on mental disorders that may affect decision-making capacity.N Engl J Med1999;340,1430-1434. [CrossRef] [PubMed]
 
Silverman, HJ Ethical considerations of ensuring an informed and autonomous consent in research involving critically ill patients.Am J Respir Crit Care Med1996;154,582-586. [PubMed]
 
Basler, JR, Martinez, EA, Winters, BD β-Adrenergic blockade accelerates conversion of postoperative supraventricular tachyarrhythmias.Anesthesiology1998;89,1052-1059. [CrossRef] [PubMed]
 
Pinder, M, Tshukutsoane, S, Scribante, J, et al Critical care research and pre-emptive informed consent: a practical approach used in Chris Hani Baragwanath ICU.Intensive Care Med1998;24,353-357. [CrossRef] [PubMed]
 
Bergner, M Patient characteristics in SUPPORT: activity status and cognitive function.J Clin Epidemiol1990;43,S33-S36. [CrossRef]
 
Brody, BA The ethics of biomedical research: an international perspective.1998,300-302 Oxford University Press. New York, NY:
 
Sackett, DL Clinical epidemiology: a basic science for clinical medicine 2nd ed.1991,69-152 Little, Brown and Company. Boston, MA:
 
Rothman, K No adjustments are needed for multiple comparisons.Epidemiology1990;1,43-46. [CrossRef] [PubMed]
 
Uhlmann, R, Pearlman, R, Cain, K Physicians’ and spouses’ predictions to elderly patients’ resuscitation preferences.J Gerontol1988;43,M115-M121. [PubMed]
 
Sulmasy, DP, Terry, PB, Weisman, CS The accuracy of substituted judgments in patients with terminal diagnoses.Ann Intern Med1998;128,621-629. [PubMed]
 
Ouslander, J, Tymchuk, A, Rahbar, B Health care decisions among elderly long term care residents and their proxies.Arch Intern Med1989;149,1367-1372. [CrossRef] [PubMed]
 
Hare, J, Pratt, C, Nelson, C Agreement between patients and their self-selected surrogates on difficult medical decisions.Arch Intern Med1992;152,1049-1054. [CrossRef] [PubMed]
 
Emanuel, E, Emanuel, L Proxy decision making for incompetent patients.JAMA1992;267,2067-2071. [CrossRef] [PubMed]
 
Sulmasy, DP, Haller, K, Terry, PB More talk, less paper: predicting the accuracy of substituted judgements.Am J Med1994;96,432-438. [CrossRef] [PubMed]
 
Epstein, LC, Lasagna, L Obtaining informed consent - form or substance?Arch Intern Med1969;123,682-688. [CrossRef] [PubMed]
 
Truog, RD, Robinson, W, Randolph, A, et al Is informed consent always necessary for randomized controlled trials?N Engl J Med1999;10,804-807
 
Etchells, E, Sharpe, G, Walsh, P Bioethics for clinicians: 1. Consent.Can Med Assoc J1996;155,177-180
 
Michels, R Are research ethics bad for our mental health?N Engl J Med1999;340,1427-1430. [CrossRef] [PubMed]
 
Gordon, N, Shade, S Advance directives are more likely among seniors asked about end-of-life care preferences.Arch Intern Med1999;159,701-704. [CrossRef] [PubMed]
 
NOTE:
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).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
[End-of-life decisions - what is important for our routine duties in critical care medicine?]. Anasthesiol Intensivmed Notfallmed Schmerzther 2012;47(3):144-9.
Guidelines
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543