Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner

Primary Care Research

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Primary care physicians may miss important patient information during initial visits with patients

Primary care doctors could do a better job of taking medical histories and improve their preventive screening skills, concludes a study supported by the Agency for Health Care Policy and Research (HS06464). Primary care physicians in this study often missed important information about a patient's related symptoms and medical history in one visit. Such omissions could adversely affect decisions about diagnostic tests and treatment plans, says Paul G. Ramsey, M.D., of the University of Washington.

Dr. Ramsey and his colleagues analyzed the history-taking and preventive screening skills of 134 primary care physicians from five Northwestern States. The researchers presented the physicians with standardized patient cases (individuals who have been trained to present a certain medical profile to the doctor) and scored them on the number of essential history and preventive screening items performed during that encounter.

About 75 percent of physicians asked questions that described presenting symptoms, and 83 percent of internists and 71 percent of family practitioners asked about medications and allergies to medications. However, only half of internists and family practitioners asked pertinent history questions related to the patient's current complaint (51 percent and 46 percent, respectively) or reviewed physical systems or past medical history (47 percent and 41 percent). For instance, 56 percent of physicians did not ask a 60-year-old man with fatigue and weight loss related to undiagnosed lymphoma whether he had night sweats, which would increase suspicion of lymphoma.

Physicians' performance in preventive screening and counseling skills also suggested opportunities for improvement. Physicians asked 74 percent of standardized patients about smoking and 60 percent about alcohol use, but they asked only 41 percent about recreational drugs. Few physicians asked about patients' sexual histories or risk factors for AIDS and other sexually transmitted diseases. The researchers suggest that physicians consider using medical intake questionnaires which could enhance medical history taking, particularly since many primary care physicians now must see more patients in less time.

See "History-taking and preventive medicine skills among primary care physicians: An assessment using standardized patients," by Dr. Ramsey, J. Randall Curtis, M.D., M.P.H., Douglas S. Paauw, M.D., and others, in the February 1998 American Journal of Medicine 104, pp. 152-158.

Today's primary care doctors offer more time and counseling to children but also prescribe more medications

Primary care physicians are spending slightly more time (14 vs. 12 minutes per visit) with young patients than they did 15 years ago, and they are providing young patients with increased preventive services and counseling. But they also are prescribing antibiotics more often (in one-third vs. one-fourth of visits) and the stimulant Ritalin 140-fold more frequently than they did 15 years ago, without clear evidence that wider use of these medications is appropriate, concludes a study supported in part by the Agency for Health Care Policy and Research (HS07892).

Timothy G. Ferris, M.D., M.Phil., of Harvard Medical School and Massachusetts General Hospital, and his colleagues used data from the National Ambulatory Care Surveys from 1979 to 1994 to analyze more than 58,000 visits by patients under the age of 18 to about 2,000 primary care physicians. During the study period, the average length of patient visits extended from 11.8 minutes in 1979 to 14.2 minutes in 1994. Longer visits were associated with Hispanic patients and the provision of counseling and preventive services such as vaccination. The time increase was less dramatic in patients belonging to health maintenance organizations (HMOs).

Physicians reported providing counseling services, including dietary and smoking counseling in 71 percent of 1994 well-child visits compared with 40 percent of 1979 visits. However, smoking cessation counseling was provided to only 2 of every 100 adolescents seen. Physicians vaccinated children in 48 percent of well-child visits in 1994 compared with 33 percent of such visits in 1980 and used blood pressure screening for children over the age of 3 in 26 percent of children in 1994 compared with 18.6 percent in 1979.

Overall prescription of antibiotics increased from 26 percent in 1978 to 31 percent of 1994 visits, adding to concerns about the growing prevalence of antibiotic-resistant bacteria. Prescriptions for methylphenidate (Ritalin) to treat attention deficit disorder increased from .01 percent of 1980 visits to 1.4 percent of 1994 visits.

See "Changes in the daily practice of primary care for children," by Dr. Ferris, Demet Saglam, M.A., Randall S. Stafford, M.D., Ph.D., and others, in the March 1998 Archives of Pediatric and Adolescent Medicine 152, pp. 227-233.

Return to Contents

Health Insurance

Managed care organizations use several techniques to discourage referral to specialists

Limiting physician referrals of patients to specialists is one of several techniques used by managed care organizations (MCOs) to reduce health care use and costs. According to a study supported by the Agency for Health Care Policy and Research (HS06833), patients in MCOs with the primary care physician (PCP) as gatekeeper are less likely to see specialists than other patients because patients typically face higher out-of-pocket costs when they self-refer. David E. Grembowski, Ph.D., of the University of Washington, and his colleagues used a social exchange theory and a review of recent studies to define the expected influence of MCOs on physician referral.

Patients who are locked in with benefits only if they see network specialists or whose networks contain a smaller percentage of specialists or have a higher ratio of PCPs to specialists probably have a lower probability of referral.

MCO utilization management programs may reduce a patient's probability of referral by requiring prior authorization of the referral by the MCO or PCP gatekeeper; approving a single specialist visit instead of multiple visits for each referral request; requiring approval of referrals by a group practice's medical director or another authority; and regular monitoring of PCP referral patterns.

MCO financial incentives also work to discourage physician referrals. The incentive to refer is greater for capitation than fee-for-service (FFS) reimbursement, which offers greater revenue for PCPs when they retain patients. In addition, referral withholds or bonus funds create a financial incentive for PCPs to refer less frequently. Finally, evidence suggests that minimal benefits with high out-of-pocket costs for a category of health services, such as mental health, decreases the probability that a patient will be referred for those services.

Details are in "Managed care and physician referral," by Dr. Grembowski, Karen Cook, Ph.D., Donald L. Patrick, Ph.D., M.S.P.H., and Amy Elizabeth Roussel, Ph.D., in the March 1998 Medical Care Research and Review 55(1), pp. 3-31.

Insurers often do not cover physician-recommended growth hormone therapy for short-stature children

Even though a physician recommends growth hormone (GH) therapy for a child, there is a good chance that it may not be covered by the child's health insurance. In fact, the contrast between physician recommendations for GH therapy and the coverage decisions of insurers is striking, according to a study supported in part by the Agency for Health Care Policy and Research (National Research Service Award training grant T32 HS00059).

Researchers from Case Western Reserve University examined the medical process influencing access to GH therapy for childhood short stature by comparing coverage policies of U.S. insurers with the treatment recommendations of U.S. physicians. Beth S. Finkelstein, Ph.D., and her colleagues mailed nationally representative surveys to 113 insurers (private, HMO, and public programs), 1,504 primary care physicians, and 534 pediatric endocrinologists.

Overall, physicians recommended GH therapy for 78 percent of children whose short stature was due to either growth hormone deficiency, Turner syndrome, or kidney failure. Of those recommended for treatment, 28 percent were denied coverage by insurers. Although 96 percent of pediatric endocrinologists recommended GH therapy for children with Turner syndrome, insurers covered GH therapy for only 52 percent of these children. Insurers would cover only 42 percent of children with chronic renal insufficiency.

GH therapy is rarely needed for life-threatening situations, is very costly (about $14,000 per year for a child weighing 20 kg), and its use has been the subject of debate. Primary care physicians are conservative in referring short children to pediatric endocrinologists who, in turn, are fairly selective in recommending children for GH treatment. Nevertheless, health insurance plans further limit access to this therapy. These findings call into question the role of insurers in final decisions about access to treatments, conclude the researchers.

See "Insurance coverage, physician recommendations, and access to emerging treatments: Growth hormone therapy for childhood short stature," by Dr. Finkelstein, J.B. Silvers, Ph.D., Ursula Marrero, M.S.S.A., and others, in the March 4, 1998, Journal of the American Medical Association 279(9), pp. 663-668.

Return to Contents

Quality of Care

Pennsylvania's pioneering effort to document quality of care for heart attack patients shows promise

Hospital scorecards based on patient mortality rates following bypass surgery and other cardiovascular procedures generate considerable publicity, with the reputations of many physicians and hospitals hanging in the balance. Given the potential impact of publicly reported scorecards, the methods used to prepare them must be analyzed critically, assert James G. Jollis, M.D., of Duke Clinical Research Institute, and Patrick S. Romano, M.D., M.P.H., of the University of California-Davis School of Medicine.

In a study supported by the Agency for Health Care Policy and Research (H08754), Drs. Jollis and Romano analyzed Pennsylvania's pioneering 1996 hospital report, Focus on Heart Attack. This report identified hospitals and physicians who had mortality rates for heart attack patients that suggested better-than-average or worse-than-average care.

According to the researchers, the Pennsylvania report has many strengths. It is exceptionally thorough and carefully explains both the potential uses and the limitations of data on outcomes. Hospitals were given the time and opportunity to ensure the accuracy of their data, appropriate techniques for model building were used, and statistical methods designed to avoid incorrect inferences due to small numbers and small differences in mortality were employed.

Despite these very positive features, the researchers did identify several significant problems, including how patients and physicians were identified, the way in which risk adjustment was performed, and the classification of deaths among patients transferred from one hospital to another. The Pennsylvania report included all patients whose hospital discharge abstracts listed a principal diagnosis of acute myocardial infarction (AMI, heart attack). It did not list patients whose principal diagnosis was a complication of AMI, such as cardiogenic shock, even if AMI was listed as a secondary diagnosis. Yet patients with such complications typically have higher mortality rates than other groups.

Also, the Pennsylvania group provided no clear guidelines for identifying patients' responsible physicians, thus making it difficult to attribute a patient's adverse outcome to the appropriate physician. The group's mortality prediction model did not distinguish whether a condition, such as congestive heart failure, was present at admission (a coexisting illness) or developed during the hospital stay (suggesting complications that may reflect poorer quality of care). This lack of distinction muddied the waters, so that expected mortality rates for some providers were probably overestimated.

Drs. Jollis and Romano describe several relatively simple ways to overcome these problems. Their suggestions include, among others, broadening the criteria used to identify AMI patients, using a consistent definition of the responsible physician, redefining risk factors based on ICD-9-CM codes to avoid adjusting for complications of care, emphasizing measures of severity that are less susceptible to manipulation by providers, and linking data from all hospitals involved in providing care for the same heart attack. Correcting these problems would probably change the current rankings of Pennsylvania hospitals and physicians, conclude the authors.

For more information, see "Pennsylvania's focus on heart attack: Grading the scorecard," by Drs. Jollis and Romano, in the April 2, 1998, New England Journal of Medicine 338(14), pp. 983-987.

Return to Contents

Legal/Ethical Issues in Medicine

Physicians' fear of malpractice suits may contribute to frequent use of PSA screening for prostate cancer

Failure to diagnose or a delay in diagnosing cancer is the most frequent allegation in medical malpractice claims in the United States. It appears that physicians who see men at risk for prostate cancer are well aware of this fact. Despite the lack of evidence that prostate specific antigen (PSA) screening causes more benefit than harm, this prostate cancer screening test is widely used by both primary care physicians (PCPs) and urologists. What's more, its use is related to physicians' malpractice concerns, according to a study by the Prostate Patient Outcomes Research Team (PORT), which is supported by the Agency for Health Care Policy and Research (HS08397).

Led by Mary McNaughton Collins, M.D., M.P.H., of Massachusetts General Hospital, the PORT researchers conducted a national survey of 429 general internists, 441 family physicians, and 586 urologists in 1995. The physicians were asked about their use of PSA screening and their interpretation of a hypothetical malpractice case of a 65-year-old man with prostate cancer who was not offered PSA screening by his PCP. The majority of PCPs reported almost always ordering PSA tests for men over age 49. Urologists were nearly unanimous about the desirability of routine PSA testing for men between ages 50 and 69, but only half recommended screening for men ages 75 to 79, and only one-quarter recommended screening for men 80 years of age and older.

About 80 percent of both PCPs and urologists thought there was some chance that the hypothetical case patient could win a malpractice suit. However, only 27 percent of PCPs and 39 percent of urologists thought the patient should win. Physicians' self-reported PSA testing practices correlated with their views on the medical malpractice vignette. Those who believed that the patient could win the suit reported routinely performing PSA testing significantly more often than those who believed the patient could not win. This suggests that medicolegal concerns are related to testing practices among this group of physicians, conclude the researchers.

Details are in "Medical malpractice implications of PSA testing for early detection of prostate cancer," by Dr. Collins, Floyd J. Fowler Jr., Ph.D., Richard G. Roberts, M.D., J.D., and others, in the Winter 1997 Journal of Law, Medicine & Ethics 25, pp. 234-242.

Genetic testing for cancer susceptibility raises ethical and psychosocial dilemmas

The development of genetic tests for susceptibility to breast, ovarian, and colon cancer has given rise to a series of ethical and psychosocial dilemmas. These issues are being examined in a series of 16 projects funded by the Cancer Genetic Studies Consortium (CGSC) of the National Institutes of Health. Benjamin S. Wilfond, M.D., supported by the Agency for Health Care Policy and Research (HS08570), and other members of the CGSC summarized these issues in a recent commentary.

The authors point out that informed consent discussions should include the potential for anxiety, altered family relationships, stigmatization, and discrimination based on test results. Strict confidentiality of the information, given insurers' access to medical records, could be problematic, and the information could alter family relationships. Those who take a test must consider whether to inform other family members of their own risk because this information also has implications for family members' cancer risk. Some adults may want their children tested, but so far, consensus is that children should not be tested for an adult-onset disease.

There are few data about how third-party payers will perceive genetic testing for cancer risks. Marketing strategies by commercial laboratories may promote testing to practitioners or directly to patients, although it is not known whether managed care organizations will pay for such tests. A more important question is whether they will fund the associated genetic counseling services.

Whether the benefits of testing outweigh the costs is a public health decision that should be made with input from those who do not have strong financial interests in the decision, says the CGSC. Although CGSC projects will provide important data about the safety and efficacy of such testing, resolution of the many issues surrounding cancer testing will require additional input from health professionals, policymakers, and the public, concludes Dr. Wilfond.

More details are in "Cancer genetic susceptibility testing: Ethical and policy implications for future research and clinical practice," by Dr. Wilfond, Karen H. Rothenberg, Elizabeth J. Thomson, Caryn Lerman, and others, in the Winter 1997 Journal of Law, Medicine & Ethics 25, pp. 243-251.

Study examines physicians' prescribing behavior

When drug companies make gifts to doctors, are such gifts really free? In a recent commentary, Anthony D. So, M.D., M.P.A., of the Agency for Health Care Policy and Research, notes that the literature suggests financial ties to drug companies may influence physicians' judgments. One study found that physicians who requested specific additions to a hospital pharmacy were substantially more likely to have met with drug company representatives or accepted money from the drug companies.

However, physicians do not seem to recognize the potential influence of drug company "gifts" on their decisionmaking. In one survey, 85 percent of medical students said it would be improper for a politician to accept $50, but only 46 percent felt it would be improper for a medical student to accept a similar contribution from a pharmaceutical company. In another study, primary care physicians maintained that they placed little weight on drug advertisements (68 percent), pharmaceutical representatives (54 percent), or patient preference (74 percent), and that they relied heavily on academic sources for drug information. Yet the study found that commercial rather than scientific sources of drug information dominated their drug choices.

For those training to become physicians, incomes are relatively low and working conditions difficult—conditions that make them particularly susceptible to the influence of drug promotion efforts. In fact, one study found that physicians in training considered gifts more influential but at the same time more appropriate than did staff physicians. It is notable, however, that ethical guidance from some professional societies makes a wider exception to their standards for medical students, residents, and fellows.

See "Free Gifts: Redundancy or Conundrum?" by Dr. So, in the March 1998 Journal of General Internal Medicine 13, pp. 213-215.

Reprints (AHCPR Publication No. 98-R045) are available from the AHCPR Publications Clearinghouse.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care