5-6503

public health initiative. We strongly sup- port this call for action, as evidence does show that neuroticism adversely affects not only the individual and his or her social environment but the generations to follow. Interventions focused on the behavioral and mental health problems associated with high neuroticism in parents need to include components addressing good parenting in order to reduce the development of endur- ing interpersonal difficulties and psychopa- thology in children of parents with high neuroticism.

REFERENCES

Belsky, J. (2008). Personality and parenting. In M. H. Bornstein (Ed.), Handbook of parenting (2nd ed., pp. 415– 438). Mahwah, NJ: Erl- baum.

Elder, G. A., Jr., Caspi, A., & Downey, G. (1986). Problem behavior and family relation- ships: Life course and intergenerational themes. In A. B. Sorensen, F. E. Weinert, & L. R. Sherrod (Eds.), Human development and the life course: Multidisciplinary perspectives (pp. 293–340). Hillsdale, NJ: Erlbaum.

Ellenbogen, M. A., & Hodgins, S. (2004). The impact of high neuroticism in parents on chil- dren’s psychosocial functioning in a popula- tion at high risk for major affective disorder: A family– environmental pathway of intergen- erational risk. Development and Psychopa- thology, 16, 113–136. doi:10.1017/ S0954579404044438

Hammen, C., Shih, J. H., & Brennan, P. A. (2004). Intergenerational transmission of de- pression: Test of an interpersonal stress model in a community sample. Journal of Consulting and Clinical Psychology, 72, 511–522. doi: 10.1037/0022-006X.72.3.511

Kochanska, G., Clark, L. A., & Goldman, M. S. (1997). Implications of mothers’ personality for their parenting and their young children’s development outcomes. Journal of Personal- ity, 65, 387– 420. doi:10.1111/j.1467- 6494.1997.tb00959.x

Lahey, B. B. (2009). Public health significance of neuroticism. American Psychologist, 64, 241–256. doi:10.1037/a0015309

Ostiguy, C., Ellenbogen, M. A., & Hodgins, S. (2009). The impact of parents’ personality on their offspring’s interpersonal functioning: A prospective 10-year study. Manuscript sub- mitted for publication.

Van Os, J., & Jones, P. B. (1999). Early risk factors and adult person– environment rela- tionships in affective disorder. Psychological Medicine, 29, 1055–1067. doi:10.1017/ S0033291799001026

Correspondence concerning this comment should be addressed to Mark A. Ellenbogen, Department of Psychology, Concordia University, 7141 Sher- brooke West, Montreal, Quebec H4B 1R6, Can- ada. E-mail: [email protected]

DOI: 10.1037/a0018051

Prescriptive Authority in the Face of Research Revelations

Roger P. Greenberg State University of New York Upstate

Medical University

Fox and his colleagues (May–June 2009) listed three occurrences beginning in the 1980s that have dampened psychologists’ desires to procure the right to prescribe psychotropic medications. One of the events noted was a review of antidepres- sant treatments that I co-authored (Green- berg & Fisher, 1989). That research review highlighted the fact that antidepressants produce a very modest effect at best when compared with placebos. Fox et al. went on to correctly note that the robust placebo effect on depression that we identified has been widely validated by subsequent stud- ies. However, Fox et al. did not point out that since the 1980s, a great deal of re- search has demonstrated even more reason to raise questions about the benefits of an- tidepressants relative to placebos and psy- chotherapy. I briefly summarize here some of the important findings that emerged from our original survey and two more recent reviews (Greenberg & Davis Gold- man, 2009; Greenberg & Fisher, 1997). Questions regarding antidepressants are of particular relevance in any discussion of prescription privileges because they are the most prescribed class of drugs in the United States (Spielmans, Thiegles, Dent, & Greenberg, 2008).

It has become evident that study de- sign is a significant factor in determining how effective antidepressants will appear to be. Overall, antidepressants look better when they are not compared with any con- trol groups (such as those receiving psy- chotherapy or taking placebos). The more a study is controlled, the less unique im- provement antidepressants seem to produce (Greenberg & Fisher, 1997). Thus, the methodology of drug trials plays a large part in how good the outcome will look.

One study feature, the double-blind design, was thought to protect results from the taint of the usual prodrug desires of investigators. This design simply assumes that neither clinicians nor patients rating outcome know whether a drug or a placebo is being administered if they are not told in advance. It turns out that investigator (and patient) bias is not held in check with this design. In double-blind studies where par- ticipants have been asked to guess what is being administered, they are able to do so

with a high degree of accuracy (Greenberg & Fisher, 1997). One possible tip-off is the fact that drugs produce significantly more side effects than placebos do. In the few studies that have used an “active placebo,” which mimics side effects (e.g., dry mouth), the modest differences between active drug and placebos shrink substan- tially.

Another reason to be concerned about the validity of proclamations regarding the therapeutic power of antidepressants rests on the finding that the published literature does not present an accurate picture of the results of effectiveness trials. Evidence has revealed that the published literature is selective and skewed in a direction that is more positive than can be justified by looking at all the conducted antidepressant trials. For example, Turner, Matthews, Linardatos, Tell, and Rosenthal (2008) presented a report looking at the Food and Drug Administration (FDA) database for all the trials— both published and unpublished— of antidepressant agents approved for use between 1987 and 2004. Only about half the trials showed placebo– drug differences. However, the published lit- erature, by not reporting on 92% of the neg- ative trials, misleads by implying that almost all the trials (94%) were positive. In addition, selective reporting deceptively suggests an effect size that is about one third larger than the one computed by using all the studies in the FDA database.

Kirsch, Moore, Scoboria, and Ni- cholls (2002, Kirsch et al., 2008) also used the FDA database to examine the benefit level of antidepressants. Looking at the six most used antidepressants approved be- tween 1987 and 1999, they concluded that antidepressants were not even 10% more effective than placebos. The outcome dif- ference on the Hamilton Depression Rating Scale was so small that they felt antidepres- sant use did not produce meaningful clini- cal change beyond that produced by place- bos. In their recent meta-analytic study, Kirsch et al. (2008) found little reason to conclude that new-generation antidepres- sants worked better than placebos for cases of depression at most levels of severity. They concluded that “given this data, there seems little evidence to support the pre- scription of antidepressant medication to any but the most severely depressed pa- tients, unless alternative treatments have failed to provide benefit” (p. 266).

Comparisons across our three major reviews (Greenberg & Davis Goldman, 2009; Greenberg & Fisher, 1989, 1997) also revealed that virtually all studies show psychotherapy is either equal to or more effective than medications for the treatment of depression. Of particular importance is

136 February–March 2010 ● American Psychologist

T hi

s do

cu m

en t i

s co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

tio n

or o

ne o

f i ts

a lli

ed p

ub lis

he rs

. T

hi s

ar tic

le is

in te

nd ed

s ol

el y

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

the fact that several studies indicate that psychotherapy provides two advantages over medications. It is more effective in preventing relapse and it produces fewer side effects. It is striking that large, sophis- ticated, recent research programs suggest that antidepressants are even less effective in “real life” than they are in controlled trials.

So, do the empirical findings touched on above lead to the conclusion that psy- chologists should not seek prescription privileges? No, they do not. However, they do lead to concerns about whether most psychologists have been briefed well enough about psychoactive drug results to give informed consent about pursuing the right to prescribe them. These concerns are bolstered by research findings showing that many of the claims made to mental health practitioners, medical professionals, and the public through journal advertisements and the media are not empirically con- firmed or carefully examined by govern- ment oversight (Donohue, Cevasco, & Rosenthal, 2007; Spielmans et al., 2008). Greater attention to medication outcome research should minimize the possibility of psychologists buying a pill in a poke.

REFERENCES

Donohue, J. M., Cevasco, M., & Rosenthal, M. B. (2007). A decade of direct-to-consumer advertising of prescription drugs. New En- gland Journal of Medicine, 357, 673– 681.

Fox, R. E., DeLeon, P. H., Newman, R., Sam- mons, M. T., Dunivin, D. L., & Baker, D. C. (2009). Prescriptive authority and psycholo- gy: A status report. American Psychologist, 64, 257–268. doi:10.1037/a0015938

Greenberg, R. P., & Davis Goldman, E. (2009). Antidepressants, psychotherapy or their com- bination: Weighing options for depression treatments. Journal of Contemporary Psycho- therapy, 39, 83–91. doi:10.1007/s10879-008- 9092-2

Greenberg, R. P., & Fisher, S. (1989). Examin- ing antidepressant effectiveness: Findings, ambiguities and some vexing puzzles. In S. Fisher & R. P. Greenberg (Eds.), The limits of biological treatments for psychological dis- tress: Comparisons with psychotherapy and placebo (pp. 1–37). Hillsdale, NJ: Erlbaum.

Greenberg, R. P., & Fisher, S. (1997). Mood- mending medicines: Probing drug, psycho- therapy, and placebo solutions. In S. Fisher & R. P. Greenberg (Eds.), From placebo to pan- acea: Putting psychiatric drugs to the test (pp. 115–172). New York, NY: Wiley.

Kirsch, I., Deacon, B. J., Heudo-Medina, T., Scoboria, A., Moore, T. J., & Johnson, B. T. (2008). Initial severity and antidepressant benefits: A meta-analysis of data submitted to the food and drug administration. PLoS Medicine, 5, 260 –268. doi:10.1371/journal .pmed.0050045

Kirsch, I., Moore, T. J., Scoboria, A., & Ni-

cholls, S. S. (2002). The emperor’s new drugs: An analysis of antidepressant medication data submitted to the U.S. Food and Drug Admin- istration. Prevention & Treatment, 5(1), Arti- cle 23. doi:10.1037/1522–3736.5.1.523a

Spielmans, G. I., Thiegles, S. A., Dent, A. L., & Greenberg, R. P. (2008). The accuracy of psy- chiatric medication advertisements in medical journals. Journal of Nervous and Mental Dis- ease, 196, 267–273.

Turner, E. H., Matthews, A. M., Linardatos, E., Tell, R. A., & Rosenthal, R. (2008). Selective publication of antidepressant trials and its in- fluence on apparent efficacy. New England Journal of Medicine, 358, 252–260.

Correspondence concerning this comment should be addressed to Roger P. Greenberg, SUNY Upstate Medical University, 750 East Adams Street, Syracuse, NY 13210. E-mail: [email protected]

DOI: 10.1037/a0018494

Effects of an Uncertain Literature on All Facets of Clinical Decision Making

Morgan T. Sammons California School of Professional

Psychology at Alliant International University

Russ Newman Alliant International University

Greenberg (2010, this issue) is correct in his assertion that the investigational heuris- tic used to measure the efficacy of antide- pressants is flawed. Robust placebo effects are endemic in the psychiatric literature, particularly in studies of antidepressants, and estimates of placebo responding have increased over time (Rief et al., 2009). The double blind, randomized clinical trial (RCT) using an inert placebo is easily bro- ken (it appears that both investigators and patients can in most instances rather quickly detect whether an active drug or a placebo is being administered), and when active placebos are used in place of inert compounds, the ability to detect a specific therapeutic drug effect diminishes (Mon- crieff, Wessely, & Hardy, 2009). This is an investigatory flaw that extends well beyond the field of psychopharmacology but is es- pecially pertinent in the investigation of any condition in which a nonspecific or subjective response (e.g., depression, pain) predominates. In the case of established treatments such as antidepressants, the pla- cebo-controlled RCT heuristic is often mis- applied (treatments of proven efficacy, no matter how modest, are arguably best un-

derstood by comparing their effects with those of other efficacious interventions, not placebos, although we recognize there is not uniformity on this position; see, e.g., Fritze & Möller, 2001).

Greenberg (2010) is also correct in noting that the psychiatric literature, per- haps more than any other corpus of inves- tigatory knowledge in medicine, has been affected by the influence of industry-spon- sored trials and industry’s control over the reporting of those trials. Author conflict of interest, often undisclosed, is common, and studies with such conflicted authorship tend to report superior results for the stud- ied drug (Lexchin, Bero, Djulbegovic, & Clark, 2003; Perlis et al., 2005). Conflict of interest appears to be frequent not only in drug trials but also in the development of clinical practice guidelines for schizophre- nia, bipolar disorder, and major depressive disorder (Cosgrove, Bursztajn, Krimsky, Anaya, & Walker, 2009). Selective pub- lishing of articles on antidepressants has been observed and may have the effect of skewing findings of meta-analytic studies of antidepressants, as such analyses rely on the published literature (Turner, Matthews, Linardatos, Tell, & Rosenthal, 2008). Sub- missions ghostwritten by drug-company- funded agents are unfortunately not diffi- cult to find in medical journals. Some (Healy, 2004) believe they are endemic in the psychiatric literature to the point that it is perhaps not unwise to suggest that all consumers of this literature (providers, re- searchers, and patients alike) should ap- proach it with caution and great skepticism.

We also agree with Greenberg’s (2010) observation that these points, how- ever valid, do nothing to gainsay the argu- ment that prescriptive authority is an ap- propriate goal for the profession of psychology. Instead, as has been our thesis, they impose a burden on us to look for optimum combinations of pharmacological and nonpharmacological interventions for mental disorders, to lessen our reliance on single interventions of proven but limited effectiveness, and to capitalize on psychol- ogy’s integrative and holistic approach to treatment. In addition, we would caution against confusing the critical need to know the limitations of any treatment used with the suggestion of an isolated application of informed consent to those psychologists who wish to pursue the right to use psy- choactive drugs as one among a number of treatments relied on.

We believe that Greenberg (2010) would likely concur with our observation that the profession of psychology, rela- tively unburdened by questionable ties to

137February–March 2010 ● American Psychologist

T hi

s do

cu m

en t i

s co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

tio n

or o

ne o

f i ts

a lli

ed p

ub lis

he rs

. T

hi s

ar tic

le is

in te

nd ed

s ol

el y

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .