Are Physicians Less Likely to Recommend Preventive Services to
Low-SES Patients?
1
Leif I. Solberg, M.D.,*
,2
Milo L. Brekke, Ph.D.,† and Thomas E. Kottke, M.D.‡
*Group Health Foundation, Minneapolis, Minnesota 55440-1309; †Brekke Associates, Minneapolis, Minnesota 55410-2252; and ‡Mayo
Clinic and Foundation, Rochester, Minnesota 55905
Background. Do low-SES adult patients visiting pri-
vate primary care clinics differ from higher SES adult
patients in their need for eight preventive services or
in receiving either a recommendation for or the
needed services?
Methods. Randomly identified adult patients were
surveyed within 2 weeks of a visit to 22 clinics in the
Minneapolis–St. Paul area. Questions assessed patient
recollection of the latest receipt of eight services and
whether needed services had been recommended dur-
ing the visit or received then soon after.
Results. Of those surveyed, 4,245 patients (1,650 low
SES) responded (84.3%), showing that low SES pa-
tients were less likely to be up to date for cholesterol
measurement, Pap smear, mammography, breast
exam, and flu or pneumonia shots (P < 0.004), but not
for blood pressure measurement. Low-SES patients
needing services received recommendations to have
them and actually received them at the same rate as
higher SES patients.
Conclusions. The 22 primary care clinics studied ap-
pear to be recommending and providing needed pre-
ventive services to visiting patients at the same rate
regardless of income or insurance status. The reasons
for differences in prevention status by SES are com-
plex but the low proportion of all patients receiving
recommendations for needed services suggests the
need to take advantage of all visits for updating pre-
vention needs. © 1997 Academic Press
Key Words: preventive medicine; health; low SES;
physician behavior.
INTRODUCTION
It is clear that populations with lower socioeconomic
status (SES) have disproportionately higher levels of
mortality and morbidity [1–8]. In fact, Pappas’ recent
review of data from several national surveys [3] and
Fein’s literature review [5] both suggest that this over-
all worse health among low-SES populations has in-
creased in the United States over the past 50 years,
and Marmot found the same for England [8].
Although both Feinstein’s and Marmot’s reviews
make clear that the reasons for this difference in
health status are very complex and poorly understood
[1,8], preventive services offer the potential to help re-
duce this problem. However, many studies have dem-
onstrated that the poor have lower rates of most pre-
ventive services [1,9–11]. That seems to be true despite
a higher frequency of many risky behaviors in this
same group [6–7,12–14]. In addition, it has also been
shown that the uninsured are only half as likely as
their insured peers to have received a variety of pre-
ventive services [15].
It has often been assumed that this lower level of
preventive services among people of lower SES is due
in part to physician behavior, i.e., that physicians are
less likely to recommend or order these services for
patients less able to afford them [16] or with other
more urgent health care priorities [17]. This assump-
tion has been fed by physician surveys that list reim-
bursement problems as a barrier to their doing a better
job of providing patients with preventive services [18].
In the course of conducting a randomized controlled
trial of the use of continuous quality improvement as a
means to improve the delivery of preventive services to
adults in primary care practices (IMPROVE) [19–20],
we have had an opportunity to gather information
about this problem. One of the specific aims of this
study is to test whether clinics in the trial will improve
their preventive services equally for their low-SES pa-
tients and their other patients.
Although we believed that the private primary care
clinics participating in the IMPROVE trial would make
systems changes that would benefit all classes of pa-
tients equally, our hypothesis was that at baseline,
low-SES patients would not be equally up to date. In
keeping with the above literature, we expected that the
level of previous preventive services as well as preven-
1
This project was supported by Grant RO1 HS08091 from the
Agency for Health Care Policy and Research
2
To whom reprint requests should be addressed at the Group
Health Foundation, 8100 34th Avenue South, P.O. Box 1309,
Minneapolis, MN 55440–1309. Fax: (612) 883–5022. E-mail:
Leif.I.Solberg@HealthPartners.com.
PREVENTIVE MEDICINE 26, 350–357 (1997)
ARTICLE NO. PM970150
350
0091-7435/97 $25.00
Copyright © 1997 by Academic Press
All rights of reproduction in any form reserved.