This is Journal Watch, a medical-literature
survey produced by the Massachusetts Medical
Society.
  Twice a week, our physician-editors summarize
important new clinical research from a group
of 25 journals. Twice a month, the summaries
get compiled into newsletter form and mailed
to subscribers.
  If you'd like to subscribe to the newsletter,
which costs $US79 a year, please call
800-843-6356; international orders:
(49) 30 335 8006.
  To comment via e-mail, please contact us at

  Contents copyright 1994, Mass. Medical Society.

Journal Watch Summaries for August 9, 1994

DOCTORS' ATTITUDES ON HEALTH CARE REFORM.
  West J Med 1994 Jul; 161:29-33.

LOVASTATIN OR NIACIN FOR HYPERCHOLESTEROLEMIA?.
  Arch Intern Med 1994 Jul 25; 154:1586-95.

THREE PAP SMEAR METHODS COMPARED.
  Obstet Gynecol 1994 Aug; 84:168-73.

GENETIC MARKERS AND PROGNOSIS OF COLORECTAL CANCER.
  N Engl J Med 1994 Jul 28; 331:213-21.

PREDICTING PSYCHOLOGICAL DYSFUNCTION AFTER TORTURE.
  JAMA 1994 Aug 3; 272:357-63.

{*filter*} QUESTIONNAIRE CAN PAVE THE WAY TO
INTERVENTION.
  J Fam Pract 1994 Jul; 39:26-32.

DOCTORS' ATTITUDES ON HEALTH CARE REFORM.
   A questionnaire survey of 1000 randomly chosen
physicians in Washington state shows why physicians are
unlikely to be unified when it comes to deciding about
health care reform.
   The response rate was 80 percent. Eighty percent of
respondents favored substantial change in health care.
Forty-three percent favored managed competition, and 40
percent a single-payer system. About two thirds felt that
malpractice reform or reducing administrative paperwork
would do "a lot" to improve the system, but only 35 percent
felt this way about a uniform benefits package for all
patients.
   Procedure-oriented physicians (surgeons, ob-gyns,
radiologists, anesthesiologists, pathologists) were more
likely than primary care physicians to favor leaving the
current system unchanged (28 vs. 8 percent). More primary-
care than procedure-oriented physicians favored a single-
payer plan (47 vs. 27 percent). Fee-for-service physicians
were more likely than salaried physicians to favor no change
(22 vs. 7 percent), and salaried physicians were more likely
to favor a single-payer system (50 vs. 34 percent). All
these subgroups were equally likely to favor managed
competition (about 45 percent).
   Comment: These physicians generally favored health-care
reform but clearly disagreed about what reform should mean.
It would have been interesting to look separately at the
opinions of salaried physicians with direct managed-care
experience. The fact that physician preferences were related
to specialty and form of remuneration is no surprise, and
these differences will probably make the political process
of reform even more difficult. --KI Marton.
   Citation: Malter AD; et al. Attitudes of Washington State
physicians toward health care reform. West J Med 1994 Jul; 161:29-33.

LOVASTATIN OR NIACIN FOR HYPERCHOLESTEROLEMIA?.
   Both lovastatin and niacin are major lipid-lowering
{*filter*}, but there are few prospective data comparing them
head-to-head. This 26-week multicenter trial compared the
two {*filter*} in 136 patients with primary hypercholesterolemia.
   The patients were put on a low-fat diet and randomized to
either lovastatin (20 mg daily, titrated to a maximum of 80
mg daily) or niacin (0.25 g three times daily, titrated to a
maximum of 4.5 g daily). More patients treated with
lovastatin than niacin reached maximum dose titration (65
vs. 53 percent). Among patients not reaching the maximum
dose who adhered to study protocol, more patients taking
lovastatin than niacin reached their treatment goals (80 vs.
32 percent); most of the remainder had dose limitations
because of adverse events or drug intolerance. Eight niacin-
treated patients stopped therapy because of side effects
(usually flushing), versus four lovastatin-treated patients.
Lovastatin, at all dose levels, was more effective than
niacin in reducing serum LDL cholesterol levels, while
niacin was more effective in increasing HDL cholesterol and
reducing triglyceride and Lp(a) lipoprotein levels.
   Comment: Both lovastatin and niacin exert dose-dependent
effects on lipids that are beneficial yet different. Thus,
the choice of these {*filter*}, used alone or in combination,
should depend upon whether lowering LDL or raising HDL (or
both) is the primary goal in a specific case. --CD Mulrow.  
   Citation: Illingworth DR; et al. Comparative effects of
lovastatin and niacin in primary hypercholesterolemia:
a prospective trial. Arch Intern Med 1994 Jul 25; 154:1586-95.
   Citation: Probstfield JL; Hunninghake DB. Nicotinic acid as a
lipoprotein-altering agent: therapy directed by the primary physician.
Arch Intern Med 1994 Jul 25; 154:1557-9.

THREE PAP SMEAR METHODS COMPARED.
   There are several techniques for obtaining
Papanicolaou smears. This prospective study evaluates three:
use of a Q-tip applicator swab plus a wooden spatula, use of
the Cytobrush plus a wooden spatula, and use of the Cervex
brush alone.
   Over 600 women attending a colposcopy clinic because of
atypical squamous cells, squamous intraepithelial lesions,
or other diagnoses were consecutively assigned to one of the
three Pap-smear techniques. All three instruments were easy
to use, caused minimal discomfort, and were acceptable to
patients. The Cytobrush caused significantly more bleeding
than the other methods; half the women assigned to it had
mild cervical bleeding. Endocervical cell clusters were best
extracted with the Cytobrush (93 percent of women had
adequate numbers), followed by the Cervex brush (68 percent)
and the swab (40 percent). The swab caused the greatest
endocervical cell distortion. All methods were similarly
sensitive in identifying histologically confirmed dysplasia.
   The authors recommend the Cytobrush as the instrument of
choice and the Q-tip swab as the least desirable because of
the devices' differing ability to obtain adequate
endocervical specimens. --CD Mulrow.
   Citation: Germain M; et al. A comparison of the three most
common Papanicolaou smear collection techniques. Obstet Gynecol
1994 Aug; 84:168-73.

GENETIC MARKERS AND PROGNOSIS OF COLORECTAL CANCER.
   The five-year survival rate in stage II colorectal
cancer (cancer extending through the bowel wall but without
node metastases) is 70 to 80 percent. Because the role of
adjuvant chemotherapy is unclear, it might be helpful to
identify the patients most likely to relapse, who may be the
best candidates for adjuvant therapy. This study explores
genetic markers as one possible prognostic approach.
   Researchers examined 135 stage II and III colorectal
tumor specimens for allelic loss of the long arm of
chromosome 18 (18q), which has been associated with
metastatic disease. Among the 65 stage II patients, those
with preserved 18q had significantly higher five-year
survival than those with loss of 18q (93 vs. 54 percent).
Among the 70 stage III patients, who had positive regional
nodes, five-year survival did not differ significantly
between the two subgroups (52 vs. 38 percent). Allelic loss
of chromosome 18q remained predictive of poor prognosis in
stage II disease after adjustment for vein invasion and
grade of tumor differentiation.  
   Comment: Genetic tumor markers like 18q certainly hold
promise as prognostic indicators in colorectal cancer. But
as editorialists point out, without randomized trials we
cannot presume that adjuvant therapy will benefit high-risk
stage II patients. --AS Brett.
   Citation: Jen J; et al. Allelic loss of chromosome 18q and
prognosis in colorectal cancer. N Engl J Med 1994 Jul 28; 331:213-21.
   Citation: Tempero M; Anderson J. Progress in colon cancer --
do molecular markers matter? N Engl J Med 1994 Jul 28; 331:267-8.

PREDICTING PSYCHOLOGICAL DYSFUNCTION AFTER TORTURE.
   Persistent psychological dysfunction is known to occur
after torture, but is not well studied. This comparison of
55 tortured political activists, 55 activists who were not
tortured, and 55 people with no history of either torture or
activism examined the effects of torture and the role of
other stressors.
   The torture victims had survived systematic torture
during detention in Turkey in the 1970s or 1980s, and most
of their accounts were validated by other sources. Post-
traumatic stress disorder (PTSD) and symptoms of anxiety and
depression were more common in the survivors than in the
other two groups. However, levels of anxiety and depression
in torture survivors were within the normal range, and the
incidence of major depression did not differ between groups.
In torture survivors, both perceived severity of torture and
a strong effect of the torture on the family predicted PTSD
symptoms; lack of family support after captivity predicted
anxiety and depression; and family history of psychiatric
illness predicted all three forms of dysfunction.
   Comment: Although small, this study suggests that the
degree of persisting psychological dysfunction after torture
depends on many variables, including factors before, during,
and after captivity. --CD Mulrow.
   Citation: Basoglu M; et al. Factors related to long-term
traumatic stress responses in survivors of torture in Turkey.
JAMA 1994 Aug 3; 272:357-63.

{*filter*} QUESTIONNAIRE CAN PAVE THE WAY TO
INTERVENTION.
   Screening for {*filter*} abuse is most effective when
coupled with intervention. In designing an intervention, it
is helpful to know the patient's own attitudes. This study
tested a questionnaire screening process designed to elicit
this information.
   Of 3750 eligible {*filter*}s from 12 urban and semirural
family practice groups in North Carolina, 2716 (72 percent)
completed the questionnaires. Those who reported any
drinking were further assessed for {*filter*} problems with the
CAGE questions, which ask whether the patient wants to Cut
down on drinking, is Annoyed by criticism about drinking,
has Guilt about drinking, or Ever drinks in the morning.
   The 53 percent of patients who reported drinking were
significantly more likely than nondrinkers to be white,
male, under age 40, and educated beyond high school. Nine
percent of drinkers (7 percent of whites and 23 percent of
blacks) answered yes to at least two CAGE questions, and 67
percent of these problem drinkers said they would like to
reduce their drinking within the next month. Health
improvement was the major motive cited for wanting to reduce
consumption; perception of {*filter*} as a stress reducer was
the most common barrier.
   Comment: This study shows that a questionnaire can
effectively screen for problem drinking, assess patients'
interest in reducing {*filter*} consumption, and identify
perceived benefits and barriers. It should be noted that all
subjects in this study were receiving regular medical care;
the questionnaire may be less useful in other populations.
--JC Puffer.
   Citation: Strecher VJ; et al. Opportunities for {*filter*} screening
and counseling in primary care. J Fam Pract 1994 Jul; 39:26-32.