The CHAIRMAN. Dr. Murray Grant is the Chief Medical Adviser for the Human Resources Division which deals with auditing of agencies and subparts of agencies involved in social services.

Dr. Grant is accompanied today by Mr. Robert Iffert, Assistant Director; Mr. Edward Murphy, auditor-economist, also of the Human Resources Division. Also accompanying Dr. Grant is Mr. James Hoffman, Supervisory Auditor of the regional office in Kansas City, Kans.

Will you please have a seat, all of you. In view of the fact that we seem always to be pressed for time, if you gentlemen have a prepared statement and can offer it for the record and then could summarize your statement for us, it would give us more time to hear all of you and more time for us to ask you questions that might be helpful.

STATEMENT OF MURRAY GRANT, M.D., CHIEF MEDICAL ADVISER, HUMAN RESOURCES DIVISION: ACCOMPANIED BY ROBERT E. IFFERT, JR., ASSISTANT DIRECTOR; EDWARD J. MURPHY, AUDITOR-ECONOMIST, HUMAN RESOURCES DIVISION; AND JAMES J. HOFFMAN, SUPERVISORY AUDITOR, GAO REGIONAL OFFICE, KANSAS CITY, KANS.

Dr. GRANT. Thank you, Mr. Chairman. I would like to introduce my colleagues who are with me here today.

Mr. Bob Iffert on my left is group director for the Human Resources Division. Beyond him is Mr. Ed Murphy, an auditor-economist with the General Accounting Office, and on my right is Mr. Jim Hoffman, the team leader for this assignment from our Kansas City office.

The CHAIRMAN. Thank you.

Dr. Grant, will you lead off?

Dr. GRANT. Mr. Chairman and members of the committee, we are pleased to be here today to discuss our report on the problems in reviews of medicaid-financed drug therapy in medicaid nursing homes which is being released today.

In summary, our report says:

"Medicaid nursing home patients taking certain classes of prescription drugs have not had their drug regimens monitored through various tests in accordance with criteria developed by Professional Standards Review Organizations (PSROS) and others. "In five of the six States visited (Georgia, Iowa, Kansas, Massachusetts and Texas), our random sample of nursing home patients indicated that 82 percent had not received the recommended tests under the most liberal available criteria for frequency. In the sixth State, California, where our sample was curtailed, a similarly high incidence of non-compliance was noted.

"Findings by PSROS evaluating long-term care in Colorado and Utah closely parallel our findings.

"To better assure the health and safety of patients, the Department of Health and Human Services in 1974 established nursing home standards requiring monthly review of patient medications by a pharmacist or registered nurse. We found that the medication review process is not working adequately. As discussed in the report, the inadequacies in the medication review process are principally due to the two following factors:

'HHS has not provided adequate information on monitoring and use of drugs, particularly the frequency with which tests should be performed; and

'HHS has not provided adequate medication review guidelines to pharmacists and nurses.'

"We also found that pharmacists making medication reviews at many of the homes we visited were also filling prescriptions for the patients. We believe this creates a potential conflict of interest because the pharmacists have a financial disincentive to reduce patient drug utilization.

"Drugs play a major role in the treatment of nursing home patients and because of the potential hazards of drug therapy, we selected for study the medication review process. Medication review involves determining whether the patient needs the drug and whether it is properly administered, effective, and safe. In medication review, the pharmacist or nurse is assisting the attending physician by assuring that the drugs are properly administered and by bringing to the physician's attention any questions regarding the effectiveness and safety of the drugs.

"The elderly, as a group, take more drugs than younger persons and elderly nursing home patients take more drugs than the noninstitutionalized elderly. A 1976 HHS report shows that 54 percent of nursing home patients were receiving six or more drugs at a time, with some receiving as many as 23 drugs. A sizable proportion of drugs prescribed for the elderly are long-term maintenance drugs, used primarily for controlling chronic diseases. The elderly generally are more susceptible to adverse drug reactions, interactions, or lack of therapeutic response because of the number of drugs they take. They are also more sensitive to some drug actions and drugs often accumulate in the body. The elderly frequently have more than one chronic medical condition which adds to the risk because a drug used to treat one condition may be contraindicated due to the presence of other conditions.

"Our approach in this review was to determine (1) whether the medications taken by selected patients were being monitored and (2) whether the general medication review procedures followed by pharmacists and registered nurses were being properly carried out. Our work was done at 68 randomly selected nursing homes in six States-California, Georgia, Iowa, Kansas, Massachusetts and Texas. The six States were selected to provide geographical dispersion and because about 30 percent of all nursing homes are located in them. Work was discontinued in California before all the homes selected were visited when it became apparent that results would not be substantially different from those in the other five States. Therefore, as explained in the report, the results from California could not be included in the statistical projections we made.

"Pharmacists and registered nurses charged with making medication reviews rely heavily on information shown on drug labeling. However, labeling generally does not have sufficient specific information on how drugs should be monitored, particularly the frequency with which tests should be performed, or on the implications of using multiple drugs affecting the same body system.

"More specific information regarding monitoring and use of some drugs is available from PSROS and other sources, but this information is not readily accessible to medication reviewers. We believe that HHS should disseminate to medication reviewers the drug

monitoring and usage criteria which has been developed by PSROS and others and promote further development of drug criteria by PSROs.

"Using the more specific information available from PSROS and other sources as criteria, we found that a high percentage of patients taking certain drugs were not receiving recommended tests under the most liberal available criteria for frequency. We also found that, in a few instances, patients either received combinations of tranquilizers or sedatives which one PSRO characterized as "inappropriate utilization" or had medical conditions which, according to drug labeling, should have precluded the use of certain drugs taken.

"We determined at each nursing home visited whether randomly selected patients taking 1 or more of 10 drugs received laboratory tests or other diagnostic tests with the frequency recommended by either selected PSRO guidelines or guidelines prepared under an HHS contract by the University of Minnesota School of Pharmacy. The Social Security Act requires PSRO's to evaluate the quality or suitability of long-term care delivered to medicaid and medicare patients.

"While many PSRO's have not developed a long-term care evaluation program, we identified five PSRO's which had developed evaluative criteria which include procedures for monitoring drugs. The HHS contracted for the guidelines developed by the University of Minnesota to assist nursing homes in identifying patterns of problems in drug utilization. Both criteria sources developed their guidelines with input from a variety of health professionals, including practicing physicians.

"The criteria represent general guidelines to be used by reviewers who are normally not physicians to identify possible problems. In those instances in which criteria are not met, the developers of the criteria intend that one or more physicians review the particular circumstances of the patient involved to determine whether deviations from the criteria were justified. This review assures that the unique health conditions and drug responses of the individual patient are taken into account.

"The 10 drugs for which we obtained monitoring criteria from PSRO's or the University of Minnesota guidelines were in one of four classes of drugs-diuretics, used to reduce body fluids; cardiac stimulants; hematinics, used to treat anemia; and antihypertensives, used to treat high blood pressure.

"The monitoring criteria specified that certain procedures, such as blood tests or electrocardiograms, be performed with various frequencies. Since not all the criteria sources recommended the same monitoring procedures or frequency of performance for the 10 drugs, we used the most lenient criteria in analyzing the sample patients. For example, we used the longest interval between tests as our standard where sources differed as to frequency of a test. "About 82 percent of the 349 patients we sampled in Georgia, Iowa, Kansas, Massachusetts, and Texas either did not receive the tests or did not receive them as frequently as recommended by the most lenient criteria available. By projecting our sample, we estimate that about 81 percent of the approximately 49,000 patients in the five States who took these drugs were not monitored as recom

mended by the criteria. Some patients took 1 or more of the 10 drugs for periods up to 26 months and did not receive the appropriate tests which generally were supposed to be performed every 6 to 12 months.

"It should be noted that those patients we identified as not meeting drug screening criteria were not reviewed by an independent physician to determine whether exceptional circumstances were present which justified deviations from criteria. On the other hand, two of the PSRO's used as criteria sources also found high rates of noncompliance with screening criteria on some of the same drugs after the added step of physician peer review showed that deviations were not justified. Because the incidence of noncompliance with the screening criteria in our sample closely parallels or is lower than the incidence of noncompliance identified by PSRO's after peer review, we believe that our findings fairly reflect the general pattern of use of these 10 drugs in nursing homes in the five States.

"We also analyzed the medications of patients randomly selected from all medicaid participants in nursing homes in the States of Georgia, Iowa, Massachusetts, and Texas. This sample did not include Kansas because we added this sample after our work in Kansas had been completed. Using prescription drug labeling information as criteria, we found that about 6 percent of the patients in our sample received contraindicated drugs, which FDA defines as "those situations in which the drug should not be used because the risk of use clearly outweighs any possible benefit." The following is an example from the cases reviewed.

"A 75-year-old woman had a recorded diagnosis of glaucoma and a recorded history of allergic reactions to sulfonamides. She did not have a recorded diagnosis of recurring urinary tract infections, although it can be inferred from other information in her medical records. Her drug regimen included a number of drugs commonly used to treat urinary tract infections.

"During a 21-month period she took Azo Gantrisin for 2 weeks, Gantrisin for 11 weeks and later again for 2 weeks, and Gantanol for a month; all three drugs are sulfonamides. She also took Urised for 2 weeks concurrently with Gantrisin. Urised is contraindicated for patients with glaucoma. We found nothing in her medical record to indicate why she was taking drugs that normally she should avoid, particularly in light of the fact that a number of other drugs which are neither sulfonamides nor contraindicated for glaucoma are available to treat urinary tract infections.

"For this same sample, we found that about 3 percent received simultaneously three or more central nervous system (CNS) depressing drugs-tranquilizers and sedatives-although one of the PSRO's guidelines indicated this was "inappropriate utilization." The following is an example of the use of multiple CNS depressing drugs selected from the cases reviewed.

"A 79-year-old woman took several different drugs that depress the CNS. Some drugs were tranquilizers or sedatives which have as a primary effect the depression of the CNS. Other drugs she took such as antihistamines depress the CNS as a side effect. At any given time, she took anywhere from one to five CNS depressing

drugs per day, although she usually took three in addition to the other drugs in her regimen.

"For a little over a year she took Triavil, a combination of an antidepressant and a tranquilizer, twice a day, and Dalmane, a nonbarbiturate sedative, once a day at bedtime. Over 6 months later, she was still taking Phenobarbital and Triavil when she also took the tranquilizers Haldol and Mellaril concurrently for a little over 2 weeks.

"We believe that most of the drug monitoring problems we identified are due to the medication reviews not having a single source of accurate, complete information on drugs. An HHS-funded study pointed out that the most widely used source of drug information by physicians is the drug labeling and we learned the same to be true for pharmacists and registered nurses making medication reviews at the homes we visited. We also found that labeling information for drugs included in our review did not always contain sufficient specific information on how the drugs should be monitored or be used in conjunction with other drugs.

"For example, labeling information for some drugs recommended that certain tests be performed but the frequency was stated in vague terms such as "frequently" or "periodically." Drug labeling for the CNS-depressing drugs we analyzed normally contained little guidance concerning the use of more than one CNS depressant, other than the admonition to use caution.

"In commenting on our draft report, HHS stated that drug labeling is intended to be used in the context of medically well-informed professionals exercising judgment concerning individual patients who have unique health conditions and drug responses and that it considered the majority of drug labeling to be adequate in that context. HHS also stated that current FDA regulations require labeling for each drug to contain information on contraindications for its use, warnings, precautions, adverse reactions, dosage and administration. According to HHS, it is impossible to include in labeling warnings regarding the use of each drug in combination with other drugs because of the myriad combinations and permutations that are possible with the number of drugs presently available.

"Although HHS stated in its comments to our draft report that drug labeling is adequate for "medically well-informed professionals, we do not believe that it is adequate for pharmacists and registered nurses making medication reviews. Studies over the last 10 years, including some sponsored by HHS, have reported that important drug information is not always readily accessible or that the information which is available, including FDA approved drug labeling, is not always complete and objective. These studies reported that health professionals, including physicians, were in need of better information than that provided in drug labeling and that one solution would be development of a drug compendium.

"In 1979, legislation was introduced in the Congress which would authorize HHS or a private organization to develop such a comprehensive drug guide. HHS is supporting this legislation although, in commenting on our draft report, it stated that it is unrealistic and possibly undesirable to expect that all information about drugs will ever be contained in one source.