LETTERS
to 15.5%). The NNT is 100/4.6=21.7. This indicates that about was a secondary outcome. Paroxetine showed a significantly
22 older patients with minor depression or dysthymia need to better effect than placebo for the primary outcome, including
change in depressive symptoms and effects on mental health
functioning. When considering all 3 outcomes, we conclude
that paroxetine showed a moderate benefit for patients with
dysthymia. Finally, we agree that the inclusion of patients re-
ceiving usual care would have yielded valuable information.
Patients in our placebo group received more visits and more
support than is typical in primary care, which may have di-
minished the observed treatment effect.
be treated with paroxetine rather than standard or placebo treat-
ment for 1 additional patient to benefit after 11 weeks of treat-
ment. The 95% CI of the NNT goes to infinity because 0 is part
of the 95% CI for the ARR. Thus, the 95% CI of the NNT (ben-
efit) is 5.68 to infinity and the NNT (harm) is 11.7 to infinity.4
Given the ARR and NNT with their 95% CIs, we believe that
paroxetine did not show benefit.
Finally, because the usual care in this group of patients is
often less extensive than the care received by patients in the
current study regardless of group, it is unfortunate that a group
receiving usual care was not included in the design. In this type
of more realistic experiment, it is possible that the magnitude
of the effect for paroxetine would have been more substantial.
John W. Williams, Jr, MD, MHS
Division of General Internal Medicine
University of Texas Health Science Center at San Antonio
1. Mynors-Wallis LM, Gath DH, Lloyd-Thomas AR, Tomlinson D. Randomised con-
trolled trial comparing problem solving treatment with amitriptyline and placebo
for major depression in primary care. BMJ. 1995;310:441-445.
2. Lynch DJ, Tamburrino MB, Nagel R. Telephone counseling for patients with
minor depression: preliminary findings in a family practice setting. J Fam Pract.
1997;44:293-298.
3. Williams JW Jr, Mulrow CD, Chiquette E, Noel PH, Aguilar C, Cornell J. A sys-
tematic review of newer pharmacotherapies for depression in adults: evidence re-
port summary. Ann Intern Med. 2000;132:743-756.
Berend Terluin, MD, PhD
Hein van Hout, PhD
Department of General Practice
Faculty of Medicine
Vrije Universiteit
Amsterdam, the Netherlands
1. Williams JW Jr, Barrett J, Oxman T, et al. Treatment of dysthymia and minor
depression in primary care: a randomized controlled trial in older adults. JAMA.
2000;284:1519-1526.
2. Cook RJ, Sackett DL. The number needed to treat: a clinically useful measure
of treatment effect. BMJ. 1995;310:452-454.
Quality of Care at Teaching
and Nonteaching Hospitals
To the Editor: Dr Allison and colleagues1 found that teach-
ing hospitals have better processes of care and outcomes for
patients with acute myocardial infarction (AMI). This is quite
different from the results of our study2 of the same Coopera-
tive Cardiovascular Project (CCP) data set. We found no in-
dependent association between teaching status and survival.
The 2 studies differ in several respects. Allison et al seem to
exclude all 39025 patients subsequently transferred to other
hospitals, whom we assigned to the initial admitting hospital.
The CCP hospitals without the ability to perform bypass sur-
gery transferred 29.9% of patients, usually for angiography and
revascularization. These patients had a 30-day mortality of 8.6%,
which is half the rate of patients who were not transferred, and
had higher compliance with process-of-care guidelines. Hos-
pitals with high levels of technology transferred only 1.9% of
their patients. Because availability of technology is strongly cor-
related with teaching status (in the study by Allison et al, 28.6%
of nonteaching vs 69.7% of major teaching hospitals had on-
site bypass surgery), the exclusion of patients who subse-
quently were transferred might explain much of the mortality
difference between nonteaching and teaching hospitals.
Allison et al included patients regardless of their preadmis-
sion status, while our study included only patients admitted
directly from home. In the CCP data, patients admitted from
nursing homes comprised 6.8% of patients at hospitals with-
out angiography vs 4.5% at hospitals with bypass surgery
(PϽ.001) and had a 30-day mortality of 41.4% compared with
18.8% for the overall CCP cohort. Conversely, patients admit-
3. Gardner SB, Winter PD, Gardner MJ, compliers. CIA (Confidence Interval Analy-
sis) software program written for use with Gardner MJ, Altman DG, eds. Statistics
With Confidence. London, England: British Medical Journal; 1989.
4. Altman DG. Confidence intervals for the number needed to treat. BMJ. 1998;
317:1309-1312.
In Reply: In response to Mr Freeny, PST-PC is a promising,
behaviorally based, psychological treatment developed specifi-
cally for primary care. In 1 randomized trial, it was superior to
placebo for major depression.1 In another trial, PST delivered
by telephone showed clinically important effects.2 We built on
previous studies by evaluating the effectiveness of PST-PC in
patients with minor depression or dysthymia, using a rela-
tively limited, but feasible dose for primary care (6 sessions).
We agree that PST-PC is best categorized as a psychological
treatment, not psychotherapy, and that the dose may have been
too low for patients with dysthymia. In an ongoing study, we
are evaluating the effect of more sessions of PST-PC. Until more
definitive data are available, we consider it a psychological treat-
ment in development.
We agree with Dr Leard-Hansson that paroxetine, like many
antidepressants, has anticholinergic and other potential ad-
verse effects. Despite this, the dropout rate due to adverse ef-
fects was only 8.7% (12/137) among the patients who received
paroxetine. This compares favorably to the average dropout rate
of 6% to 11% due to adverse effects that we cited.3 Although we
agree that medications with strong anticholinergic properties
should be avoided in patients with cognitive impairment, par-
oxetine is certainly not contraindicated in elderly patients.
Drs Terluin and van Hout’s calculations of ARR and NNT ted from outpatient clinics comprised 8.5% of patients at hos-
are correct but only consider remission from depression, which pitals with low levels of technology vs 10.1% at hospitals with
2994 JAMA, December 20, 2000—Vol 284, No. 23 (Reprinted)
©2000 American Medical Association. All rights reserved.