PHARMACOEPIDEMIOLOGY Edited by Abraham G. Hartzema, Miquel S. Porta, and Hugh H. Tilson
ANTIBIOTIC USE IN URBAN WlllTES AND BLACKS: THE MINNESOTA HEART SURVEY Sandra L. Melnick, J. Michael Sprafka, David L. Laitinen, Roberd M. Bostick, John M. Flack, and Gregory L. Burke
To assess racial differences in the use of antibiotics, including penicillins,erythromycins,tetracyclines,sulfas, and cephalosporins. DESIGN: Population-based surveys, conducted from 1985 to 1987.
OBJECTIVE:
SETIlNG: The seven-countymetropolitanarea of Minneapolis-St. Paul,MN. PARTICIPANTS: 3127 whites (response rate 68 percent) and 1047 blacks (response rate 65 percent), aged 35-74 years. RESULTS: White women (26 percent of 1625) were more likely to report having taken an antibiotic in the past year than were white men (18 percent of 1502), black women (18 percent of 590), or black men (15 percent of 457). Reported antibiotic usage decreased with increasingage. Black men were more likely than white men to report the use of tetracyclinesor sulfas; otherwise, white men reported higher usage prevalences.White women reported higher usage prevalencesof all drug classes than black women.
Significantindependent predictors of antibiotic use were younger age, white race, and female gender. Potential explanationsfor these differences include differences in patient access,physician-prescribing behaviors, or both.
CONCLUSIONS:
Ann Pharmacother 1992;26:1292-5. SANDRA L. MELNICK, Dr.P.H., was with the Division of Epidemiology, School of Public Health, University of Minnesota; she is now an Assistant Professor, National Institute of Allergy and Infectious Diseases, Division of AIDS, Rockville, MD; J, MICHAEL SPRAFKA, Ph.D., is an Associate Professor, Division of Epidemiology, School of Public Health, University of Minnesota, 1300 S. Second St., Suite 300, Minneapolis, MN 55454; DAVID L. LAITINEN, M.P.H., is a Statistician, ParkNicollet Medical Foundation, Minneapolis; ROBERD M. BOSTICK, M.D., is an Assistant Professor, Division of Epidemiology, School of Public Health, University of Minnesota; JOHN M. FLACK, M.D., is an Assistant Professor, Division of Epidemiology, School of Public Health and Division of General Internal Medicine, University of Minnesota; and GREGORY L. BURKE, M.D., is an Associate Professor, Department of Public Health Sciences, Bowman Gray School of Medicine, WinstonSalem, NC. Reprints: J. Michael Sprafka, Ph.D. This study was supported by National Heart, Lung, and Blood Institute Grant ROl23727. ABRAHAM G. HARTZEMA, M.S.P.H., Pharm.D., Ph.D., is a Professor, Pharmacy Administration, School of Pharmacy, and a Clinical Associate Professor, Health Policy and Administration, School of Public Health, University of North Carolina, Chapel Hill, NC 27599; MIQUEL S. PORTA, M.D., M.P.H., is an Adjunct Associate Professor, Department of Epidemiology, School of Public Health, University of North Carolina; an Associate Professor, Universitat Autonorna de Barcelona; and the Director, Clinical Epidemiology Unit, Institut Municipal d'Investigaci6 Medica, Barcelona, Spain; and HUGH H. TILSON, M.D., M.P.H., Dr.P.H., is the Director, Division of Epidemiology, Surveillance, and Pharmacoeconomics, Burroughs Wellcome Company; a Clinical Professor, Department of Family Medicine, School of Medicine, and an Adjunct Professor, Department of Epidemiology, School of Public Health, and Department of Pharmacy Administration, School of Pharmacy, University of North Carolina, Chapel Hill, NC 27599.
1292
•
nonwhite patient populations have been noted to be less likely than physicians with predominantly white patient populations to recommend preventive measures, such as screening mammography or influenza immunization. Practice disparities have been ascribed to differences in physician training, the socioeconomic status of patients, and the time spent with each patient. I Less well-documented are possible racial differences in patients' use of prescription drugs. Differences observed may be attributable to patient characteristics such as the method of healthcare reimbursement and income; differences in the prevalence of treatment-seeking behaviors among patients, disease patterns, patient practices, or physician treatment practices. In 1984, systemic antibiotics were the second most commonly prescribed drug class, with the number of outpatient prescriptions being dispensed remaining relatively stable during the period 1975-1984.2 Because of the high prescription rate and stability, as well as the fact that they are not available overthe-counter, systemic antibiotics are useful indicators of racial differences in patterns of medication use. The purpose of this study was to test the hypothesis that there are differences between whites and blacks with regard to the frequency and type of antibiotics used. Antibiotic drug information was obtained by patient self-report in a population-based survey of white and black adults living in the Minneapolis-St. Paul (Twin Cities) metropolitan area. PHYSICIANS WITH PREDOMINANTLY
Methods STUDY POPULATION
Methods and findings from the Minnesota HeartSurvey, designed 10 collect repeated measures of coronary heart disease risk factors from probability samplesof residents of the Twin Citiesmetropolitan area in 1980-82 and againin 1985-87, have beendescribed.Y Briefly, the seven-county metropolitan area was divided into 704 clusters of approximately 1000 households each,basedon 1970 and 1980 census data. From these, 40 clusters and approximately ten percent of the households within each clusterwere randomly selected. One individual aged 25-74 years was then randomly selectedfrom each household. Excluded were 1322 peopleaged 25-34 years, 101 blacks, 4 peoplewithpermanent residence elsewhere, and 25 who were too ill or unableto communicate. In addition, a surveyof blacksresiding in 4.5 contiguous censustractsin Minneapolis was conductedbetweenJuly and November 1985, to provide moreprecise estimates of theircoronary heartdisease riskfactors. Every occupiedhousing unit in the sampling area was identified and all black
The Annalsof Pharmacotherapy • 1992 October, Volume 26
residents aged 35-74 years were recruited into the survey.v This report compares antibiotic use among white participants from the second survey period (1985-87) with antibiotic use among blacks in the 1985 survey. CORONARY HEART DISEASE RISK FACTORS The Minnesota Heart Survey consisted of a home interview followed by a clinic visit. The home interview elicited information on gender, age, income (dichotomized for this report as < $15 OOO/year vs. ~$15 OOO/year), education (:,> high school vs. > high school), employment (unemployed vs. employed), marital status (currently unmarried vs. currently married), cigarette smoking status (current smoker vs. all others), alcohol use (drinking ~I ounce per week vs. drinking
ANTIBIOTIC USE IN URBAN WlllTES AND BLACKS: THE MINNESOTA HEART SURVEY Sandra L. Melnick, J. Michael Sprafka, David L. Laitinen, Roberd M. Bostick, John M. Flack, and Gregory L. Burke
To assess racial differences in the use of antibiotics, including penicillins,erythromycins,tetracyclines,sulfas, and cephalosporins. DESIGN: Population-based surveys, conducted from 1985 to 1987.
OBJECTIVE:
SETIlNG: The seven-countymetropolitanarea of Minneapolis-St. Paul,MN. PARTICIPANTS: 3127 whites (response rate 68 percent) and 1047 blacks (response rate 65 percent), aged 35-74 years. RESULTS: White women (26 percent of 1625) were more likely to report having taken an antibiotic in the past year than were white men (18 percent of 1502), black women (18 percent of 590), or black men (15 percent of 457). Reported antibiotic usage decreased with increasingage. Black men were more likely than white men to report the use of tetracyclinesor sulfas; otherwise, white men reported higher usage prevalences.White women reported higher usage prevalencesof all drug classes than black women.
Significantindependent predictors of antibiotic use were younger age, white race, and female gender. Potential explanationsfor these differences include differences in patient access,physician-prescribing behaviors, or both.
CONCLUSIONS:
Ann Pharmacother 1992;26:1292-5. SANDRA L. MELNICK, Dr.P.H., was with the Division of Epidemiology, School of Public Health, University of Minnesota; she is now an Assistant Professor, National Institute of Allergy and Infectious Diseases, Division of AIDS, Rockville, MD; J, MICHAEL SPRAFKA, Ph.D., is an Associate Professor, Division of Epidemiology, School of Public Health, University of Minnesota, 1300 S. Second St., Suite 300, Minneapolis, MN 55454; DAVID L. LAITINEN, M.P.H., is a Statistician, ParkNicollet Medical Foundation, Minneapolis; ROBERD M. BOSTICK, M.D., is an Assistant Professor, Division of Epidemiology, School of Public Health, University of Minnesota; JOHN M. FLACK, M.D., is an Assistant Professor, Division of Epidemiology, School of Public Health and Division of General Internal Medicine, University of Minnesota; and GREGORY L. BURKE, M.D., is an Associate Professor, Department of Public Health Sciences, Bowman Gray School of Medicine, WinstonSalem, NC. Reprints: J. Michael Sprafka, Ph.D. This study was supported by National Heart, Lung, and Blood Institute Grant ROl23727. ABRAHAM G. HARTZEMA, M.S.P.H., Pharm.D., Ph.D., is a Professor, Pharmacy Administration, School of Pharmacy, and a Clinical Associate Professor, Health Policy and Administration, School of Public Health, University of North Carolina, Chapel Hill, NC 27599; MIQUEL S. PORTA, M.D., M.P.H., is an Adjunct Associate Professor, Department of Epidemiology, School of Public Health, University of North Carolina; an Associate Professor, Universitat Autonorna de Barcelona; and the Director, Clinical Epidemiology Unit, Institut Municipal d'Investigaci6 Medica, Barcelona, Spain; and HUGH H. TILSON, M.D., M.P.H., Dr.P.H., is the Director, Division of Epidemiology, Surveillance, and Pharmacoeconomics, Burroughs Wellcome Company; a Clinical Professor, Department of Family Medicine, School of Medicine, and an Adjunct Professor, Department of Epidemiology, School of Public Health, and Department of Pharmacy Administration, School of Pharmacy, University of North Carolina, Chapel Hill, NC 27599.
1292
•
nonwhite patient populations have been noted to be less likely than physicians with predominantly white patient populations to recommend preventive measures, such as screening mammography or influenza immunization. Practice disparities have been ascribed to differences in physician training, the socioeconomic status of patients, and the time spent with each patient. I Less well-documented are possible racial differences in patients' use of prescription drugs. Differences observed may be attributable to patient characteristics such as the method of healthcare reimbursement and income; differences in the prevalence of treatment-seeking behaviors among patients, disease patterns, patient practices, or physician treatment practices. In 1984, systemic antibiotics were the second most commonly prescribed drug class, with the number of outpatient prescriptions being dispensed remaining relatively stable during the period 1975-1984.2 Because of the high prescription rate and stability, as well as the fact that they are not available overthe-counter, systemic antibiotics are useful indicators of racial differences in patterns of medication use. The purpose of this study was to test the hypothesis that there are differences between whites and blacks with regard to the frequency and type of antibiotics used. Antibiotic drug information was obtained by patient self-report in a population-based survey of white and black adults living in the Minneapolis-St. Paul (Twin Cities) metropolitan area. PHYSICIANS WITH PREDOMINANTLY
Methods STUDY POPULATION
Methods and findings from the Minnesota HeartSurvey, designed 10 collect repeated measures of coronary heart disease risk factors from probability samplesof residents of the Twin Citiesmetropolitan area in 1980-82 and againin 1985-87, have beendescribed.Y Briefly, the seven-county metropolitan area was divided into 704 clusters of approximately 1000 households each,basedon 1970 and 1980 census data. From these, 40 clusters and approximately ten percent of the households within each clusterwere randomly selected. One individual aged 25-74 years was then randomly selectedfrom each household. Excluded were 1322 peopleaged 25-34 years, 101 blacks, 4 peoplewithpermanent residence elsewhere, and 25 who were too ill or unableto communicate. In addition, a surveyof blacksresiding in 4.5 contiguous censustractsin Minneapolis was conductedbetweenJuly and November 1985, to provide moreprecise estimates of theircoronary heartdisease riskfactors. Every occupiedhousing unit in the sampling area was identified and all black
The Annalsof Pharmacotherapy • 1992 October, Volume 26
residents aged 35-74 years were recruited into the survey.v This report compares antibiotic use among white participants from the second survey period (1985-87) with antibiotic use among blacks in the 1985 survey. CORONARY HEART DISEASE RISK FACTORS The Minnesota Heart Survey consisted of a home interview followed by a clinic visit. The home interview elicited information on gender, age, income (dichotomized for this report as < $15 OOO/year vs. ~$15 OOO/year), education (:,> high school vs. > high school), employment (unemployed vs. employed), marital status (currently unmarried vs. currently married), cigarette smoking status (current smoker vs. all others), alcohol use (drinking ~I ounce per week vs. drinking
