Letters
with no events in 1 or both groups, as was the case in our study with 9 groups having no mortality events. The Peto method has also been shown to be effective for studies with numerically balanced groups and moderate treatment effect sizes, as in our meta-analysis.2 Bradford and colleagues also express concerns regarding our use of a fixed-effects model that may not assume adequate heterogeneity. Using the random-effects DerSimonian-Laird model to calculate the pooled effect estimates would have produced greater bias with multiple zero-event groups1 and probable similar overall effect estimates, as the heterogeneity assessed for mortality events in our study was negligible at I2 = 15% (P = .28). In addition, our Bayesian random-effects analysis was largely congruent with the frequentist analyses. Further validating the robustness of our conclusions, a recent analysis from an independent group of researchers3 using a random-effects Mantel-Haenszel method with continuity correction for largely the same set of trials garnered results nearly identical to ours. Bradford and colleagues request an interaction P value to evaluate the difference between subgroups of patients older than 65 years vs those 65 years or younger. The calculated z score based on the published event rates was 1.79, translating into a statistically significant (P < .10 for interaction) difference in bleeding rates (P = .07 for interaction).4 We agree that an ecological fallacy remains a possible source of bias for this interaction term, given the lack of access to individual patient-level data in trials performed over the past several decades. Dr Bova expresses concerns about smaller, older studies driving the results of our primary analysis.5 To avoid selection bias, all randomized trials meeting our prespecified inclusion criteria were included in our primary analysis. However, our results proved consistent in sensitivity analyses examining only the 6 most recent trials performed over the past 5 years as well as among the 8 trials examining patients with intermediate-risk pulmonary embolism. Trial sequential analyses of our end points also confirmed that the primary findings are unlikely to change with further accumulated data. Although the PEITHO trial5 was the largest trial of thrombolysis in pulmonary embolism, it was underpowered to evaluate mortality. We do not think that our results argue for routine use of thrombolysis in all patients with intermediate-risk pulmonary embolism. We believe this decision remains predicated on individual clinical circumstances and agree that more sophisticated risk stratification tools could help delineate optimal candidates for thrombolytic treatment. Saurav Chatterjee, MD Jay Giri, MD, MPH Author Affiliations: Division of Cardiology, St Luke’s-Roosevelt Hospital, New York, New York (Chatterjee); Cardiovascular Medicine Division, University of Pennsylvania Perelman School of Medicine, Philadelphia (Giri).
1590
1. Bradburn MJ, Deeks JJ, Berlin JA, Russell Localio A. Much ado about nothing: a comparison of the performance of meta-analytical methods with rare events. Stat Med. 2007;26(1):53-77. 2. Cochrane Collaboration. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions, version 5.1.0 [updated March 2011]. http://www.cochrane-handbook.org. Accessed December 20, 2013. 3. Marti C, John G, Konstantinides S, et al. Systemic thrombolytic therapy for acute pulmonary embolism: a systematic review and meta-analysis [published online June 10, 2014]. Eur Heart J. doi:10.1093/eurheartj/ehu218. 4. Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ. 2003;326(7382):219. 5. Meyer G, Vicaut E, Danays T, et al; PEITHO Investigators. Fibrinolysis for patients with intermediate-risk pulmonary embolism. N Engl J Med. 2014;370 (15):1402-1411.
Recurrence of Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis To the Editor Dr Finkelstein and colleagues1 used hospital discharge data to determine the incidence and risk of recurrence of Stevens-Johnson syndrome and toxic epidermal necrolysis, which are rare serious conditions. The authors did not review medical records. With expert review, only about 80% of prospectively ascertained cases diagnosed as Stevens-Johnson syndrome or toxic epidermal necrolysis had these diagnoses confirmed.2 The incidence of Stevens-Johnson syndrome and toxic epidermal necrolysis and the proportion of cases diagnosed as StevensJohnson syndrome in this study (750 hospitalized cases/13 million people over 10 years or 5.8 million person-years) are higher than in studies that included expert case review (0.4-1.2 per 1 million person-years for toxic epidermal necrolysis and 1-6 per 1 million person-years for Stevens-Johnson syndrome).2 Finkelstein and colleagues noted a “several thousand-fold higher” recurrence risk in those previously diagnosed with Stevens-Johnson syndrome or toxic epidermal necrolysis. This observation may reflect cases of erythema multiforme and other skin eruptions often due to infectious agents such as herpes simplex that often reoccur and are misdiagnosed as Stevens-Johnson syndrome, as well as cases of Stevens-Johnson syndrome and toxic epidermal necrolysis due to causes other than drugs. About 15% of cases in adults are due to causes other than drugs.3 A prior pediatric case series noted recurrence in 10 of 55 patients with Stevens-Johnson syndrome or toxic epidermal necrolysis.4 Only 2 of 13 recurrences were attributed to drugs, and the cause was unknown for 3 of 13.4 Risk of recurrence with reexposure to related drugs is high, but only about a 4-fold increase in risk of hypersensitivity to unrelated drugs is seen in patients with a history of drug allergy.5 Avoiding reexposure to both causative and related drugs in persons who have experienced Stevens-Johnson syndrome or toxic epidermal necrolysis is essential. However, most recurrent cases of Stevens-Johnson syndrome or toxic epidermal necrolysis are due to causes other than drugs. Robert S. Stern, MD
Corresponding Author: Jay Giri, MD, MPH, Cardiovascular Medicine Division, University of Pennsylvania Perelman School of Medicine, 3400 Spruce St, Philadelphia, PA 19104 ([email protected]).
Author Affiliation: Department of Dermatology, Beth Israel Deaconess Medical Center, Boston, Massachusetts.
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Corresponding Author: Robert S. Stern, MD, Department of Dermatology, Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA 02215 (rstern @bidmc.harvard.edu).
JAMA October 15, 2014 Volume 312, Number 15
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a RITSUMEIKAN UNIVERSITY GROUP User on 06/07/2015
jama.com
Letters
Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported serving as a consultant in legal matters for law firms representing McNeil Laboratories and Eisei Pharma; and serving as a consultant to Bristol-Myers Squibb and Boeringer-Ingelheim. All activities were related to issues concerning adverse cutaneous reactions to medications. 1. Finkelstein Y, Macdonald EM, Li P, Hutson JR, Juurlink DN. Recurrence and mortality following severe cutaneous adverse reactions. JAMA. 2014;311(21): 2231-2232. 2. Roujeau JC, Kelly JP, Naldi L, et al. Medication use and the risk of Stevens-Johnson syndrome or toxic epidermal necrolysis. N Engl J Med. 1995; 333(24):1600-1607.
undoubtedly heterogeneous (urticaria, anaphylaxis, delayed hypersensitivity, etc). This estimate has no bearing on the risk of recurrence in patients with documented Stevens-Johnson syndrome or toxic epidermal necrolysis. We reiterate our assertion that patients with a first episode of Stevens-Johnson syndrome or toxic epidermal necrolysis are at high risk of recurrence, and that clinicians should use care when prescribing medications to this vulnerable population.
3. Sassolas B, Haddad C, Mockenhaupt M, et al. ALDEN, an algorithm for assessment of drug causality in Stevens-Johnson Syndrome and toxic epidermal necrolysis: comparison with case-control analysis. Clin Pharmacol Ther. 2010;88 (1):60-68.
Yaron Finkelstein, MD, ABCP(Dip) Erin M. Macdonald, MSc David N. Juurlink, MD, PhD
4. Finkelstein Y, Soon GS, Acuna P, et al. Recurrence and outcomes of Stevens-Johnson syndrome and toxic epidermal necrolysis in children. Pediatrics. 2011;128(4):723-728.
Author Affiliations: University of Toronto, Toronto, Ontario, Canada (Finkelstein); Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada (Macdonald); Sunnybrook Research Institute, Toronto, Ontario, Canada (Juurlink).
5. Strom BL, Schinnar R, Apter AJ, et al. Absence of cross-reactivity between sulfonamide antibiotics and sulfonamide nonantibiotics. N Engl J Med. 2003; 349(17):1628-1635.
In Reply Stevens-Johnson syndrome and toxic epidermal necrolysis are rare disorders. To study their recurrence over time, large data sets are needed. This precludes a case-bycase examination of medical records. Dr Stern suggests that erythema multiforme and other less severe conditions may have been miscoded as StevensJohnson syndrome and toxic epidermal necrolysis. A 1995 study by Roujeau et al 1 identified miscoding in 16% of cases. However, in our study we used the most recent version of the International Statistical Classification of Diseases, Tenth Revision codes, which clearly differentiates Stevens-Johnson syndrome (code L51.1) and toxic epidermal necrolysis (code L51.2) from various forms of erythema multiforme (codes L51.0, L51.8, L51.9). The number of cases of Stevens-Johnson syndrome and toxic epidermal necrolysis in our sample (708 cases among approximately 13 million people over 9 years, or 6 cases per 1 million person-years; Stevens-Johnson syndrome, 4.8 per 1 million person-years; toxic epidermal necrolysis, 1.2 cases per 1 million person-years) is in accord with the baseline rate suggested by previous studies, including the study by Roujeau et al.1 In addition, the high mortality rate (18%) among our patients is consistent with Stevens-Johnson syndrome and toxic epidermal necrolysis but not mild exanthema. Even though some cases of Stevens-Johnson syndrome may be unrelated to drugs, the medical literature suggests at least 85% are induced by a drug, whereas toxic epidermal necrolysis is almost exclusively induced by a drug.2 A number of studies3-5 have previously shown that drug exposure is a common scenario leading to Stevens-Johnson syndrome recurrence. Stern states that a 3-fold increase in hypersensitivity to unrelated drugs is seen in patients with a history of drug allergy. This figure derives from a database study6 of 96 patients with sulfonamide allergy, in whom the nature of both the first and subsequent adverse drug reactions was not reported but was jama.com
Corresponding Author: Yaron Finkelstein, MD, ABCP(Dip), Hospital for Sick Children, University of Toronto, 555 University Ave, Toronto, ON M5G 1X8, Canada ([email protected]). Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported. 1. Roujeau JC, Kelly JP, Naldi L, et al. Medication use and the risk of Stevens-Johnson syndrome or toxic epidermal necrolysis. N Engl J Med. 1995; 333(24):1600-1607. 2. Mockenhaupt M. The current understanding of Stevens-Johnson syndrome and toxic epidermal necrolysis. Expert Rev Clin Immunol. 2011;7(6):803-813. 3. Finkelstein Y, Soon GS, Acuna P, et al. Recurrence and outcomes of Stevens-Johnson syndrome and toxic epidermal necrolysis in children. Pediatrics. 2011;128(4):723-728. 4. Brown M, Yowler C, Brandt C. Recurrent toxic epidermal necrolysis secondary to iopromide contrast. J Burn Care Res. 2013;34(1):e53-e56. 5. Huang L-Y, Liao W-C, Chiou C-C, Lou J-P, Hu P, Ko F-C. Fatal toxic epidermal necrolysis induced by carbamazepine treatment in a patient who previously had carbamazepine-induced Stevens-Johnson syndrome. J Formos Med Assoc. 2007;106(12):1032-1037. 6. Strom BL, Schinnar R, Apter AJ, et al. Absence of cross-reactivity between sulfonamide antibiotics and sulfonamide nonantibiotics. N Engl J Med. 2003; 349(17):1628-1635.
Commitment Devices to Improve Unhealthy Behaviors To the Editor A Viewpoint by Dr Rogers and colleagues1 discussed the underuse of commitment devices to improve unhealthy behaviors. The authors underlined 2 basic features of these devices: their voluntary use by patients who want to change behaviors and their reinforcement of consequences related to failed goal achievement by patients. One of the ways that Rogers et al1 proposed to increase the uptake of these devices was by requiring patients to opt out of commitment devices. Even though I support nudge interventions to modify population health, I foresee some barriers or precautions to take when generalizing findings from studies on commitment devices. First, interventions on the default options in which participation is assumed unless individuals opt out have been shown effective,2 whereas studies on commitment devices are based on voluntary participation, thereby involving participants who are willing to at least consider a behavioral change. However, a JAMA October 15, 2014 Volume 312, Number 15
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a RITSUMEIKAN UNIVERSITY GROUP User on 06/07/2015
1591
with no events in 1 or both groups, as was the case in our study with 9 groups having no mortality events. The Peto method has also been shown to be effective for studies with numerically balanced groups and moderate treatment effect sizes, as in our meta-analysis.2 Bradford and colleagues also express concerns regarding our use of a fixed-effects model that may not assume adequate heterogeneity. Using the random-effects DerSimonian-Laird model to calculate the pooled effect estimates would have produced greater bias with multiple zero-event groups1 and probable similar overall effect estimates, as the heterogeneity assessed for mortality events in our study was negligible at I2 = 15% (P = .28). In addition, our Bayesian random-effects analysis was largely congruent with the frequentist analyses. Further validating the robustness of our conclusions, a recent analysis from an independent group of researchers3 using a random-effects Mantel-Haenszel method with continuity correction for largely the same set of trials garnered results nearly identical to ours. Bradford and colleagues request an interaction P value to evaluate the difference between subgroups of patients older than 65 years vs those 65 years or younger. The calculated z score based on the published event rates was 1.79, translating into a statistically significant (P < .10 for interaction) difference in bleeding rates (P = .07 for interaction).4 We agree that an ecological fallacy remains a possible source of bias for this interaction term, given the lack of access to individual patient-level data in trials performed over the past several decades. Dr Bova expresses concerns about smaller, older studies driving the results of our primary analysis.5 To avoid selection bias, all randomized trials meeting our prespecified inclusion criteria were included in our primary analysis. However, our results proved consistent in sensitivity analyses examining only the 6 most recent trials performed over the past 5 years as well as among the 8 trials examining patients with intermediate-risk pulmonary embolism. Trial sequential analyses of our end points also confirmed that the primary findings are unlikely to change with further accumulated data. Although the PEITHO trial5 was the largest trial of thrombolysis in pulmonary embolism, it was underpowered to evaluate mortality. We do not think that our results argue for routine use of thrombolysis in all patients with intermediate-risk pulmonary embolism. We believe this decision remains predicated on individual clinical circumstances and agree that more sophisticated risk stratification tools could help delineate optimal candidates for thrombolytic treatment. Saurav Chatterjee, MD Jay Giri, MD, MPH Author Affiliations: Division of Cardiology, St Luke’s-Roosevelt Hospital, New York, New York (Chatterjee); Cardiovascular Medicine Division, University of Pennsylvania Perelman School of Medicine, Philadelphia (Giri).
1590
1. Bradburn MJ, Deeks JJ, Berlin JA, Russell Localio A. Much ado about nothing: a comparison of the performance of meta-analytical methods with rare events. Stat Med. 2007;26(1):53-77. 2. Cochrane Collaboration. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions, version 5.1.0 [updated March 2011]. http://www.cochrane-handbook.org. Accessed December 20, 2013. 3. Marti C, John G, Konstantinides S, et al. Systemic thrombolytic therapy for acute pulmonary embolism: a systematic review and meta-analysis [published online June 10, 2014]. Eur Heart J. doi:10.1093/eurheartj/ehu218. 4. Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ. 2003;326(7382):219. 5. Meyer G, Vicaut E, Danays T, et al; PEITHO Investigators. Fibrinolysis for patients with intermediate-risk pulmonary embolism. N Engl J Med. 2014;370 (15):1402-1411.
Recurrence of Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis To the Editor Dr Finkelstein and colleagues1 used hospital discharge data to determine the incidence and risk of recurrence of Stevens-Johnson syndrome and toxic epidermal necrolysis, which are rare serious conditions. The authors did not review medical records. With expert review, only about 80% of prospectively ascertained cases diagnosed as Stevens-Johnson syndrome or toxic epidermal necrolysis had these diagnoses confirmed.2 The incidence of Stevens-Johnson syndrome and toxic epidermal necrolysis and the proportion of cases diagnosed as StevensJohnson syndrome in this study (750 hospitalized cases/13 million people over 10 years or 5.8 million person-years) are higher than in studies that included expert case review (0.4-1.2 per 1 million person-years for toxic epidermal necrolysis and 1-6 per 1 million person-years for Stevens-Johnson syndrome).2 Finkelstein and colleagues noted a “several thousand-fold higher” recurrence risk in those previously diagnosed with Stevens-Johnson syndrome or toxic epidermal necrolysis. This observation may reflect cases of erythema multiforme and other skin eruptions often due to infectious agents such as herpes simplex that often reoccur and are misdiagnosed as Stevens-Johnson syndrome, as well as cases of Stevens-Johnson syndrome and toxic epidermal necrolysis due to causes other than drugs. About 15% of cases in adults are due to causes other than drugs.3 A prior pediatric case series noted recurrence in 10 of 55 patients with Stevens-Johnson syndrome or toxic epidermal necrolysis.4 Only 2 of 13 recurrences were attributed to drugs, and the cause was unknown for 3 of 13.4 Risk of recurrence with reexposure to related drugs is high, but only about a 4-fold increase in risk of hypersensitivity to unrelated drugs is seen in patients with a history of drug allergy.5 Avoiding reexposure to both causative and related drugs in persons who have experienced Stevens-Johnson syndrome or toxic epidermal necrolysis is essential. However, most recurrent cases of Stevens-Johnson syndrome or toxic epidermal necrolysis are due to causes other than drugs. Robert S. Stern, MD
Corresponding Author: Jay Giri, MD, MPH, Cardiovascular Medicine Division, University of Pennsylvania Perelman School of Medicine, 3400 Spruce St, Philadelphia, PA 19104 ([email protected]).
Author Affiliation: Department of Dermatology, Beth Israel Deaconess Medical Center, Boston, Massachusetts.
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Corresponding Author: Robert S. Stern, MD, Department of Dermatology, Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA 02215 (rstern @bidmc.harvard.edu).
JAMA October 15, 2014 Volume 312, Number 15
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a RITSUMEIKAN UNIVERSITY GROUP User on 06/07/2015
jama.com
Letters
Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported serving as a consultant in legal matters for law firms representing McNeil Laboratories and Eisei Pharma; and serving as a consultant to Bristol-Myers Squibb and Boeringer-Ingelheim. All activities were related to issues concerning adverse cutaneous reactions to medications. 1. Finkelstein Y, Macdonald EM, Li P, Hutson JR, Juurlink DN. Recurrence and mortality following severe cutaneous adverse reactions. JAMA. 2014;311(21): 2231-2232. 2. Roujeau JC, Kelly JP, Naldi L, et al. Medication use and the risk of Stevens-Johnson syndrome or toxic epidermal necrolysis. N Engl J Med. 1995; 333(24):1600-1607.
undoubtedly heterogeneous (urticaria, anaphylaxis, delayed hypersensitivity, etc). This estimate has no bearing on the risk of recurrence in patients with documented Stevens-Johnson syndrome or toxic epidermal necrolysis. We reiterate our assertion that patients with a first episode of Stevens-Johnson syndrome or toxic epidermal necrolysis are at high risk of recurrence, and that clinicians should use care when prescribing medications to this vulnerable population.
3. Sassolas B, Haddad C, Mockenhaupt M, et al. ALDEN, an algorithm for assessment of drug causality in Stevens-Johnson Syndrome and toxic epidermal necrolysis: comparison with case-control analysis. Clin Pharmacol Ther. 2010;88 (1):60-68.
Yaron Finkelstein, MD, ABCP(Dip) Erin M. Macdonald, MSc David N. Juurlink, MD, PhD
4. Finkelstein Y, Soon GS, Acuna P, et al. Recurrence and outcomes of Stevens-Johnson syndrome and toxic epidermal necrolysis in children. Pediatrics. 2011;128(4):723-728.
Author Affiliations: University of Toronto, Toronto, Ontario, Canada (Finkelstein); Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada (Macdonald); Sunnybrook Research Institute, Toronto, Ontario, Canada (Juurlink).
5. Strom BL, Schinnar R, Apter AJ, et al. Absence of cross-reactivity between sulfonamide antibiotics and sulfonamide nonantibiotics. N Engl J Med. 2003; 349(17):1628-1635.
In Reply Stevens-Johnson syndrome and toxic epidermal necrolysis are rare disorders. To study their recurrence over time, large data sets are needed. This precludes a case-bycase examination of medical records. Dr Stern suggests that erythema multiforme and other less severe conditions may have been miscoded as StevensJohnson syndrome and toxic epidermal necrolysis. A 1995 study by Roujeau et al 1 identified miscoding in 16% of cases. However, in our study we used the most recent version of the International Statistical Classification of Diseases, Tenth Revision codes, which clearly differentiates Stevens-Johnson syndrome (code L51.1) and toxic epidermal necrolysis (code L51.2) from various forms of erythema multiforme (codes L51.0, L51.8, L51.9). The number of cases of Stevens-Johnson syndrome and toxic epidermal necrolysis in our sample (708 cases among approximately 13 million people over 9 years, or 6 cases per 1 million person-years; Stevens-Johnson syndrome, 4.8 per 1 million person-years; toxic epidermal necrolysis, 1.2 cases per 1 million person-years) is in accord with the baseline rate suggested by previous studies, including the study by Roujeau et al.1 In addition, the high mortality rate (18%) among our patients is consistent with Stevens-Johnson syndrome and toxic epidermal necrolysis but not mild exanthema. Even though some cases of Stevens-Johnson syndrome may be unrelated to drugs, the medical literature suggests at least 85% are induced by a drug, whereas toxic epidermal necrolysis is almost exclusively induced by a drug.2 A number of studies3-5 have previously shown that drug exposure is a common scenario leading to Stevens-Johnson syndrome recurrence. Stern states that a 3-fold increase in hypersensitivity to unrelated drugs is seen in patients with a history of drug allergy. This figure derives from a database study6 of 96 patients with sulfonamide allergy, in whom the nature of both the first and subsequent adverse drug reactions was not reported but was jama.com
Corresponding Author: Yaron Finkelstein, MD, ABCP(Dip), Hospital for Sick Children, University of Toronto, 555 University Ave, Toronto, ON M5G 1X8, Canada ([email protected]). Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported. 1. Roujeau JC, Kelly JP, Naldi L, et al. Medication use and the risk of Stevens-Johnson syndrome or toxic epidermal necrolysis. N Engl J Med. 1995; 333(24):1600-1607. 2. Mockenhaupt M. The current understanding of Stevens-Johnson syndrome and toxic epidermal necrolysis. Expert Rev Clin Immunol. 2011;7(6):803-813. 3. Finkelstein Y, Soon GS, Acuna P, et al. Recurrence and outcomes of Stevens-Johnson syndrome and toxic epidermal necrolysis in children. Pediatrics. 2011;128(4):723-728. 4. Brown M, Yowler C, Brandt C. Recurrent toxic epidermal necrolysis secondary to iopromide contrast. J Burn Care Res. 2013;34(1):e53-e56. 5. Huang L-Y, Liao W-C, Chiou C-C, Lou J-P, Hu P, Ko F-C. Fatal toxic epidermal necrolysis induced by carbamazepine treatment in a patient who previously had carbamazepine-induced Stevens-Johnson syndrome. J Formos Med Assoc. 2007;106(12):1032-1037. 6. Strom BL, Schinnar R, Apter AJ, et al. Absence of cross-reactivity between sulfonamide antibiotics and sulfonamide nonantibiotics. N Engl J Med. 2003; 349(17):1628-1635.
Commitment Devices to Improve Unhealthy Behaviors To the Editor A Viewpoint by Dr Rogers and colleagues1 discussed the underuse of commitment devices to improve unhealthy behaviors. The authors underlined 2 basic features of these devices: their voluntary use by patients who want to change behaviors and their reinforcement of consequences related to failed goal achievement by patients. One of the ways that Rogers et al1 proposed to increase the uptake of these devices was by requiring patients to opt out of commitment devices. Even though I support nudge interventions to modify population health, I foresee some barriers or precautions to take when generalizing findings from studies on commitment devices. First, interventions on the default options in which participation is assumed unless individuals opt out have been shown effective,2 whereas studies on commitment devices are based on voluntary participation, thereby involving participants who are willing to at least consider a behavioral change. However, a JAMA October 15, 2014 Volume 312, Number 15
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a RITSUMEIKAN UNIVERSITY GROUP User on 06/07/2015
1591
