Secondary Logo

Journal Logo

Article

Accuracy of surrogate decision making in elective surgery

Mantravadi, Anand V. MD; Sheth, Bhavna P. MD; Gonnering, Russell S. MD; Covert, Douglas J. MD, MPH

Author Information
Journal of Cataract & Refractive Surgery: December 2007 - Volume 33 - Issue 12 - p 2091-2097
doi: 10.1016/j.jcrs.2007.07.036
  • Free

Abstract

For competent and capable patients, autonomous medical decision making is clearly the ethical standard.1–4 With a rising life expectancy comes an increase in the prevalence of dementia. Improved resuscitation and life-sustaining measures may also come at the price of significant neurological compromise. Thus, the decision-making capacity of a growing segment of the population is compromised. Advanced directives, living wills, and surrogate decision making are efforts to preserve patients' autonomy in situations in which they are no longer able to make their own decisions. Advanced directives and living wills are designed to maintain the patient's voice, enabling predetermined documentation to influence his or her course of health care. A surrogate (proxy) is defined as a person acting on behalf of another, or a substitute (http://dictionary.law.com. Accessed August 10, 2007). The theoretical accuracy of surrogate decision making is based on a “substituted judgment standard”4,5 that uses the patient's beliefs, values, and best interests as the foundation for medical decision making.

Mounting evidence, clinical experience, and several well-publicized cases have shown that these current methods of medical decision-making are significantly inaccurate. Surrogate decision makers often project their own values and beliefs into decision making,6 have low accuracy in predicting patient wishes,7 and often have high false-positive rates.8 Moreover, family members appear to be poor predictors of patients' wishes,9 overestimating the amount of medical intervention patients would want, while physicians tend to underestimate.4 Seckler et al.4 found that the concordance of resuscitation preferences of elderly patients with their chosen surrogates and physicians was no better than chance alone in the accurate prediction of the patients' wishes. These findings, focused largely on end-of-life scenarios, have significant implications for current standards of medical decision making.

Cataract surgery is the most common elective surgical procedure performed, increasing at an estimated rate of 5% annually. This study assessed the accuracy of the current standard of surrogate decision making in the previously unexplored arena of elective cataract surgery. Using a hypothetical scenario of the presence of a cataract in the current state of health and in a state of progressive dementia, the preferences of patients, their designated surrogates, and ophthalmologists were evaluated.

PATIENTS AND METHODS

Thirty-seven patients meeting inclusion criteria were identified during a visit (new or follow-up) to a single provider in a comprehensive ophthalmology clinic of a large teaching hospital. Institutional review board approval of the study protocol was obtained before patient enrollment. All new patients and routine annual follow-up patients during predetermined scheduled clinics were approached for enrollment. Postoperative patients were not eligible for the study.

Criteria for entry included English language facility, written consent to participate, Folstein Mini-Mental Status Score higher than 20,10 and absence of significant eye noncataractous pathology (eg, advanced glaucoma, macular degeneration) as the etiology of vision loss. The Folstein Mini-Mental status examination, a rapid, commonly used method of assessing cognitive mental status, is often used as a research screening tool.11 Performance on the Folstein Mini-Mental status examination varies by age and education; in this study, a score of higher than 20 was used, as in a study by Seckler et al.,4 because patients scoring above 20 are often asked to make their own health-care decisions.

Before examination, patients were asked to designate a family member or friend whom they preferred as a surrogate decision maker should medical decisions need to be made on their behalf. Patients were provided a written brief description of cataract and cataract surgery (Appendix A) and were asked to note their preferences on a Likert scale in 2 mental status scenarios: current state of health and moderate dementia (Appendix B). Patients were advised that their designated surrogate would be contacted and were asked not to discuss the details of the study with the surrogate. Surrogates were contacted by a single investigator in person or by telephone within 48 hours of the patient's office visit. Surrogates were read a brief description of the study purpose and provided verbal consent. Identical descriptions of cataract and cataract surgery were read, and surrogates were asked to predict the patient's preferences for surgery under circumstances of the patient's current state of health and hypothetical moderate dementia (Appendix C). Community ophthalmologists were independently asked to predict a hypothetical patient's wishes for surgery in the current state of health and dementia scenarios and asked to assess their own and the surrogates' accuracy in decision making (Appendix D).

The percentage agreement between the surrogate's predictions and the patient's preferences was determined. The degree of concordance between the surrogate's predictions and the patient's preferences was determined using the κ coefficient for dichotomous Likert-scale data and the chi-square analysis for concordance beyond that expected of chance. For the κ coefficient analysis, a reduced model was used in which an “uncertain” answer was assumed to be equivalent to a “no” for dichotomy. The κ coefficient ranges from 0 to 1, where 0 is level of chance agreement and 1 is perfect agreement. A κ coefficient greater than 0.4 is required to conclude that a moderate or greater degree of agreement exists beyond that which can be expected due to chance alone.4,12 A full chi-square model is so termed as each of the 5 choices on a Likert scale (yes, probably yes, uncertain, probably not, no) were incorporated into the analysis.

RESULTS

Of the 37 patients enrolled in the study (Table 1), 32 had designated surrogate decision makers who could be contacted within 48 hours for paired analyses. All patients met criteria for entry, and 59% had a living will or advanced directive.

Table 1
Table 1:
Patient characteristics.

In the current state of health scenario, with a visually significant cataract, most patients desired cataract surgery (94%), while physicians projected all such patients would want surgery. In a scenario of progressive dementia with a cataract, 50% of the patients desired cataract surgery and 25% of physicians thought the patients would want surgery.

Although all patients attested to understanding the role of a surrogate decision maker in medical care, 94% of surrogates and 75% of physicians indicated the same understanding.

Figure 1 shows the paired data between the patients and their designated surrogate decision makers in the current state of health with a visually significant cataract based on question 1 to the patient and surrogate (Appendix B and Appendix C, respectively). The percentage agreement between the patients and their designated surrogate decision maker (5-choice scale) was 72%. As determined by the reduced models of κ coefficient and full model of chi-square analysis, in the patient's current state of health, there was more agreement than that expected by chance alone between the patients and surrogate decision makers (κ = 0.652, P<.0001; chi square = 16.5, P<.002).

Figure 1
Figure 1:
Current state of health scenario. Designated surrogate decision maker's understanding of patient's elective-surgery preference.

Figure 2 shows the paired data between patients and their designated surrogate decision makers in a state of progressive dementia with a visually significant cataract based on question 2 to the patient and surrogate (Appendix B and Appendix C, respectively). The percentage agreement between the patients and their designated surrogate decision makers (5-choice scale) was 34%. Using the reduced model κ coefficient and the full model chi-square analysis, there was no more agreement than expected by chance alone (κ = .228, P = .19; chi square = 14.4, P = .57).

Figure 2
Figure 2:
Progressive dementia scenario. Designated surrogate's understanding of patient's elective-surgery preference.

Analysis of the paired data between patients and their designated surrogates (question 6 in Appendix B and Appendix C, respectively) showed disagreement about whether a discussion regarding a patient's preferences had even occurred (Figure 3, percentage agreement 53%; chi square = 0.43, P = .84). Thus, when patients and surrogates were asked independently whether elective surgical preferences were discussed should a future situation arise causing inability of autonomous decision making, nearly half the patient–surrogate pairs were in disagreement that such a discussion took place.

Figure 3
Figure 3:
Discussion of elective surgery situations between the patient and designated surrogate.

Table 2 shows an analysis of responses from questions 3 and 4 (Appendices B to D), exploring the level of accuracy patients felt of their designated surrogates and physicians and the level of accuracy surrogates and physicians felt of each other in representing a patient's wishes. Although 91% of patients felt their designated surrogates to be accurate representatives, fewer patients (72%) felt their physicians were accurate decision makers on their behalf. In addition, although 78% of physicians felt surrogates were accurate representatives for their patients, only 56% of physicians felt as confident in their own accuracy of surrogate decision making for patients. Table 3 shows the professed understanding of the role of a surrogate decision maker in elective surgery by most patients and surrogates (100% and 94% respectively). In contrast, 75% of physicians professed such understanding.

Table 2
Table 2:
Assessment of accuracy of decision making.
Table 3
Table 3:
Patient, proxy, and physician comprehension of the surrogate decision maker's role.

DISCUSSION

Previous studies of the accuracy of the current approach of medical decision making for mentally incapacitated patients for end-of-life and resuscitation preferences have raised several concerns about the validity of this standard approach.4,8–9,13–16 Using hypothetical scenarios, these studies found poor concordance and astonishingly low levels of agreement between patients' wishes and those predicted by their proxies and physicians.

This study sought to assess the accuracy of the currently implemented approach to medical decision making in the relatively unexplored arena of elective surgery, using cataract surgery as a representative elective procedure. When a hypothetical scenario of progressive dementia was introduced, 97% of designated proxies felt their decisions were accurate; however, proxies were no better than chance in accurately predicting the wishes of patients. Disagreement about whether a discussion of preferences even took place is particularly disturbing given our reliance on this method for decision making, illustrating that the confidence in the accuracy of proxies may be misguided.

Only 56% of physicians felt their ability to predict a patient's wishes in a scenario of moderate dementia was accurate. Furthermore, although all physicians predicted patients would want cataract surgery in their current state of health, only 25% felt patients would want surgery in the moderate dementia scenario. This finding in elective surgery is consistent with the suggestion put forth by Seckler et al.4 that family members tend to err on the side of providing an intervention while physicians tend to err on the side of withholding an intervention in reference to resuscitation preferences.

The disturbing implications of the inaccuracy of surrogate decision making for end-of-life scenarios and resuscitation preferences are self-evident. The decision to proceed with elective cataract surgery by patients, proxies, and physicians in scenarios of moderate dementia may bear less gravity—likely supported by the widely held public perceptions of rapid visual rehabilitation and success rates. However, the critical ethical importance of accurately approximating incapacitated patients' true wishes should be held to the highest standards regardless of the proposed medical intervention.

The best corrected visual acuity of patients enrolled in the study, although part of the ensuing clinical evaluation, was not a parameter used in this study. Although patients were randomly recruited, it is possible that if more patients had poor visual acuity, both the patients and their designated surrogates might be more inclined to prefer cataract surgery. It is also possible that if more patients had good vision, understanding cataract-related vision loss might be more difficult and they might be less inclined to prefer cataract surgery. However, as this study was based on hypothetical scenarios, the fundamental premises relied on an assumption that patients can extract themselves from their current state and imagine a projected scenario. This is one inherent weakness of hypothetical scenarios. The biases of the designated surrogates based on their own visual status cannot be eliminated. For example, surrogates who have had successful or complicated cataract surgery might be more or less inclined, respectively, to recommend such a procedure for a patient unable to make his or her own decisions if that patient has an advanced cataract. But this bias closely mimics what would actually occur in practice when attempting to rely on a surrogate for medical decision making on behalf of a patient.

Another weakness of hypothetical situations is the applicability to real-life situations.4 However, to assess accuracy of such a process, hypothetical scenarios are inevitable, and advanced directives and living wills are by definition hypothetical.4,15 Other weaknesses of this study are the small number of patient–proxy pairs, limited clinical setting (single outpatient provider clinic), and inability to truly assess a patient's or proxy's comprehension of the nature of cataract surgery. Comprehension of the nature of surgery (risks, benefits, alternatives) is a problem that draws on the larger underlying problem with true informed consent, which applies to all surgery.

The participants were asked whether they understood the role of a decision maker without previously provided information of expected roles simply to gain insights regarding a “perceived understanding,” which more likely approximates a real-life scenario of decision making for elective surgery. Despite professed understanding, the accuracy of the decision making was not enhanced. We believe this is a product of the very principle upon which this kind of decision making is founded. People are asked to make decisions for others without a real understanding of their role, the concepts of substituted judgment, and the inability to limit personal biases. These parameters (understanding/comprehension) are very challenging to quantify objectively. Patients and surrogates were not given more information by the interviewer than the physicians were. It is interesting that more patients and surrogates have a professed understanding than physicians regarding the role of a decision maker. It is possible that the training of physicians incorporates a sensitivity to issues of patient autonomy, particularly in the arena of elective surgery; therefore, making surgical decisions on behalf of a hypothetical patient who cannot make decisions for himself or herself is perhaps a source of uneasiness.

The fundamental basis for surrogate decision making is the well-intentioned propagation of patient autonomy in situations of mental incapacity.17 The inaccuracy of such an approach in elective surgery, as suggested in this study, along with previous findings in end-of-life scenarios, represents a complex ethical dilemma. Clearly, surrogate decision making is a valuable approach to decision making. However, the inaccuracies of such projections suggest that a renewed emphasis on “best interest” considerations by all parties involved and advanced written plans with specific attention to elective surgical scenarios may enhance this decision-making process.

REFERENCES

1. United States. President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Making health care decisions. 1982. U.S. Government Printing Office. Washington, DC. chapter VI. Available at: https://idea.iupui.edu/dspace/bitstream/1805/961/1/Making%20health%20care%20decisions%20-%20Chapter%206.pdf. Accessed August 10, 2007.
2. Evans DA, Funkenstein HH, Albert MS, et al. Prevalence of Alzheimer's disease in a community population of older persons. Higher than previously reported. JAMA. 1989;262:2551-2556.
3. Schoenberg BS, Kokmen E, Okazaki H. Alzheimer's disease and other dementing illnesses in a defined United States population: incidence rates and clinical features. Ann Neurol. 1987;22:724-729.
4. Seckler AB, Meier DE, Mulvihill M, Paris BEC. Substituted judgment: how accurate are proxy predictions? Ann Intern Med. 1991;115:92-98.
5. United States. President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Deciding to forego life sustaining treatment. 1983. U.S. Government Printing Office. Washington, DC. 132–133. Available at: http://www.bioethics.gov/reports/past_commissions/deciding_to_forego_tx.pdf. Accessed August 10, 2007.
6. Fagerlin A, Ditto PH, Danks JH, et al. Projection in surrogate decisions about life-sustaining medical treatments. Health Psychol. 2001;20:166-175.
7. Coppolino M, Ackerson L. Do surrogate decision makers provide accurate consent for intensive care research? Chest. 119. 2001. 603-612. Available at: http://www.chestjournal.org/cgi/reprint/119/2/603. Accessed August 10, 2007.
8. Ouslander JG, Tymchuk AJ, Rahbar B. Health care decisions among elderly long-term care residents and their potential proxies. Arch Intern Med. 1989;149:1367-1372.
9. Zweibel NR, Cassel CK. Treatment choices at the end of life: a comparison of decisions by older patients and their physician-selected proxies. Gerontologist. 1989;29:615-621.
10. Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.
11. Crum RM, Anthony JC, Bassett SS, Folstein MF. Population-based norms for the Mini-Mental State Examination by age and educational level. JAMA. 1993;269:2386-2391.
12. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159-174.
13. Uhlmann RF, Pearlman RA, Cain KC. Physicians and spouses predictions of elderly patients' resuscitation preferences. J Gerontol. 1988;43(5):M115-M121.
14. Uhlmann RF, Pearlman RA, Cain KC. Understanding of elderly patients' resuscitation preferences by physicians and nurses. West J Med. 1989;150:705-707.
15. Tomlinson T, Howe K, Notman M, Rossmiller D. An empirical study of proxy consent for elderly persons. Gerontologist. 1990;30:54-64.
16. Diamond EL, Jernigan JA, Moseley RA, et al. Decision-making ability and advance directive preferences in nursing home patients and proxies. Gerontologist. 1989;29:622-626.
17. Bramstedt KA. Questioning the decision-making capacity of surrogates. Int Med J. 2003;33:257-259.

APPENDIX A

Cataract Surgery

(Adapted from the American Academy of Ophthalmology patient brochure on Cataract Surgery, 2004. Available at: https://secure3.aao.org/pdf/051112.pdf. Accessed August 10, 2007.)

What Is a Cataract?

A cataract is a clouding of the eye's naturally clear lens. The lens focuses light rays on the retina—the layer of light-sensing cells lining the back of the eye—to produces a sharp image of what we see. When the lens becomes cloudy, light rays cannot pass through it easily, and vision is blurred.

What Causes Cataracts?

Cataract development is a normal process of aging, but cataracts also develop from eye injuries, certain diseases, or medications; even genes may play a role.

How Can a Cataract Be Treated?

A cataract may not need to be treated if your vision is only slightly blurry. Surgery is the only way to remove a cataract. Cataract surgery can be considered when you are no longer able to see well enough to do the things you like to do.

In cataract surgery, the cloudy lens is removed from the eye through a surgical incision. The natural lens is replaced with a permanent intraocular lens.

The Day of Surgery

Surgery is usually outpatient, involving some sedation to make you comfortable, and local anesthesia. Under an operating microscope, an incision is made in the eye, instruments are inserted to remove the cloudy lens, and a synthetic lens is implanted.

After Surgery

Drops prescribed by your doctor will have to be used, and the doctor will have to see you again for follow-up visits after surgery.

The success rate of cataract surgery is excellent, and improved vision is achieved in the majority of patients if other vision-limiting problems are not present. If other eye diseases, such as macular degeneration (aging changes in the retina), glaucoma, or diabetic retinopathy, are present, some patients may not see as well as they would like to, even if the cataract surgery is successful.

Complications After Cataract Surgery

Although rare, serious complications after cataract surgery can occur, including but not limited to: (1) infection, (2) bleeding, (3) swelling of the retina, (4) detachment of the retina, (5) drooping of the eyelid, and (6) the need for further surgery.

APPENDIX B

Patient Questionnaire

  1. Imagine yourself in your current state of health. If decreased vision due to a cataract began to impact the quality of your life, would you want to have cataract surgery?
  2. YesProbably yesUncertainProbably notNo
  3. Now imagine yourself with memory loss and senility. Imagine you were unable to care for yourself and make medical decisions. If a significant cataract were noted during ophthalmic (eye) exam – would you want to have cataract surgery?
  4. YesProbably yesUncertainProbably notNo
  5. How accurate do you think your family member/decision maker will be at representing your wishes in elective surgery?
  6. AccurateFairly accurateUncertainNot too accurateInaccurate
  7. How accurate do you think your eye physician will be at representing your wishes in elective surgery?
  8. AccurateFairly accurateUncertainNot too accurateInaccurate
  9. How well do you understand the role of a surrogate decision maker?
  10. Understand fullyMostlyPartiallyDon't understand at all
  11. Have you discussed elective surgery situations when decision-making capacity is impaired with your family?
  12. YesNo
  13. Do you have a living will, advanced directive, or other written document outlining your wishes regarding any end-of-life health issues such as but not limited to
  14. (a) the use of ventilators (breathing machines)?
  15. (b) resuscitation preferences (CPR: the use of chest compressions to maintain circulation, electric paddles to shock the heart if it stopped, etc.)?
    • YesNo

APPENDIX C

Designated Surrogate Questionnaire

After reading the question, circle the answer you feel is most appropriate.

  1. Imagine your family member in current state of health. If decreased vision due to a cataract began to impact the quality of his or her life, do you think he or she would want to have cataract surgery?
  2. YesProbably yesUncertainProbably notNo
  3. Imagine your family member with memory loss and senility and that he or she were unable to care for himself or herself and make medical decisions. If a significant cataract were noted during ophthalmic (eye) examination, would your family member want to have cataract surgery?
  4. YesProbably yesUncertainProbably notNo
  5. Do you think your prediction/decision is an accurate representation of your family member's wishes in elective surgery?
  6. AccurateFairly accurateUncertainNot too accurateInaccurate
  7. How accurate do you think your eye physician will be at representing your family member's wishes in elective surgery?
  8. AccurateFairly accurateUncertainNot too accurateInaccurate
  9. How well do you understand the role of a surrogate decision maker?
  10. Understand fullyMostlyPartiallyDon't understand at all
  11. Have you discussed elective surgery situations when decision-making capacity is impaired with your family member?
  12. YesNo

APPENDIX D

Ophthalmologist Questionnaire

Please circle answers.

  1. Imagine a patient in a decent state of health. If decreased vision due to a cataract began to impact the quality of his or her life, do you think he or she would want to have cataract surgery?
  2. YesProbably yesUncertainProbably notNo
  3. Imagine your same patient with memory loss and senility—unable to care for himself or herself and make medical decisions. If a significant cataract were noted during the ophthalmic examination, would your patient want to have cataract surgery?
  4. YesProbably yesUncertainProbably notNo
  5. Do you think your prediction/decision is an accurate representation of your patient's wishes in elective surgery?
  6. AccurateFairly accurateUncertainNot too accurateInaccurate
  7. How accurate do you think a surrogate decision maker (designated family member, friend, or representative) will be at representing your patient's wishes for cataract surgery?
  8. AccurateFairly accurateUncertainNot too accurateInaccurate
  9. How well do you understand the role of a surrogate decision maker for medical decisions?
  10. Understand fullyMostlyPartiallyDon't understand at all
© 2007 by Lippincott Williams & Wilkins, Inc.