Pharmacist instruction and patient compliance with medication therapy

This study was to determine the effect of a pharmacists' instruction on the ability of one hundred people to take their medication as their physician prescribed it. These people were patrons of a retail pharmacy, and patients of private physicians. Pour groups of twenty five were selected, and each was given a different type of instruction about their medication. The first group was given no instruction. The second group was given oral instruction. The third group was given oral and written instruction. The fourth group was given oral instruction, and had their prescription labeled with a time of day dose schedule. The rate of error was found by determining the ratio of errors to the number of doses the patient should have taken on the fifth day after a prescription was filled. The number of errors was determined by visiting the patient at his home, and counting the doses remaining in the prescription container. There was a decrease in error from 26% in group one to 16% in group two, and 17% in group three and 10„9% in group four. The results of this study indicate a need for instruction to the patient about his medications, and the effect of four types of instruction be a pharmacist.

PHARMAC 1ST 11'~STRUCTI('1\; AND? ATIEN:£I C(IMPLIAl~CE WITH MEJJICATIGN THERAPY by KAY HIS1(;P ALLEN A thesis submitted to the faculty of the University of Utah in partial fulfillment of the reQuirements for the degree of - Master of Science in Hospital Pharmacy Department of Applied Pharmaceutical 3ciences University of Utah March 1974 SUI' E R\/ J :~ C) R L of a tht'::ls subrp.!tted by dt[:l<:.e. . D::tc .. ~ '--'-~ FI I'.~' AL ,\ P P ){ () "V A L DI l-:.;" tbe SIlpcnri:-.:}::,)' kk I"CPU: c:tations ~u. . (l ;)1 "lti;rrc..tiR~ m(.4tt'ri:ds nr: ~li If';<:n uscri fJ ~ 1S Appro'.'Nl f-·,< the Gradu"tC' CC:1.!z:cil t\ .I, I:, I ,~ir .1'f:':. ~~~~ tj~ l'~-ccn ~ ACKNOWLEDGMENTS I would like to acknowledge the participation and support of the following people. First the customers who so willingly became part of the study. Second the physicians who gave their support and suggestions. Third the supervisory committe: Professor Jan N.Bair for his guidance and encouragement, Dr. Gary Gregor, Dr. James D. McMahon, and Dr. Robert V. Petersen for their suggestions and interest in the study. I am particularly appreciative of the interest and assurance given me by my beloved wife Mae. TABLE () F ACKN0WLEDGMENTS • • LIST OF TABLES . LIST uF FIGURES .. ':,- ABSTRACT .. .. .. .. . . · .. . . . · .. . .. .. .. ·. .. • • .. II c() NTENT S • .. .. .. .. .. .. . .. .. • .. .. .. .. . · . . .. . iv . · . . . .. " vii .. .. .. .. .. .. .. ix .. .. .. .. .. . .. .. .. .. .. .. x .. .. .. .. Chapter 1. INTR(;DUCTI0N • • • .. .. STATEMENT GF THE PRCBLEM BASIC ASSUMPTI0NS .. .. .. .. • .. 0 .. • .. .. 1 • .. 0 .. • • .. 0 .. • • 5 7 .... 0 ... 0 ........ 0 .. 0 .. .. .. .. 0 0 .. .. 8 2. DESIGN uF STUDY • • .. .. .. .. • • .. .. .. • .. .. • • 10 DELIMITATIGNS CRITERIA . 0 .. • • • 0 0 • • • .. .. .. .. 10 • • • 13 • • • .. • 0 .. .. • .. • .. .. • COL L~C T IG N AND ANALYSIS CF DATA 0 .. .. .. .. • • .. .. .. • • .. • 3. RESULTS UF THE STUDY .. 0 15 • • • 0 • • .. • .. • .. • • • • • .. • 15 Group une • • • • • • .. • .. • • • • .. • • 0 15 Group Two • • • • • • • .. .. • • • • .. .. .. 0 21 Group Three .. • 0 .. • • .. • .. • .. .. • .. • • 21 Group Four • o 0 .. .. .. GROUP DATA eLM? ARIS(/N U~' THE GROUPS . . .. . . • · . . . . .. . . . . . • • • .. 0 Comparison of Group Two to Group One • .. .. Comparison of Group Three to Group 0ne • .. Comparison of Group Three to Group Two .. .. Comparison of Group Four to Group One • • 21 22 22 .. 30 . 33 0 33 Comparison of Group Four to Groups Two and Three • • • • • • • • • • • 0 34 4. DISCUSSIGN 0F THE FINDINGS • • • • • • • • • • 37 ANALYSIS OF THE PRCJBLEMS • • • • • • • • • • RELATIuNS WITH PATRC)NS OF THE PHA..1:UIIACY • • • • • 37 • • 46 Contact at Home Interview 41 • • • • • • • • 47 Repeat Contacts •• 0 0 0 • • • • • • • • 48 PHYSICIAN H.ESPGNSE ." e • • • • • • • • 0 • 48 STRUCTun.AL ELEltENTS (:li' PHARMACY ROUTINE EFFECT OF STUDY ON PH_tJtMACY EUSI:NzSS 5. SUMl'IIARY • • THE PRCBLEM METHOD • • • 0 • • 49 • • • 0 • .. • • • • • • • • • 0 51 • • • • • • • • • • • • • • CI • • 51 0 • • • • • • • • • • • • • • • • 51 ·. ............. • 131131 I() GHA?HY · . .. . APPENDIX •• • • • • .. VITA • • • 49 • FINDINGS • • • • ceNCL USI0 NS • • • • • • • • • • 52 52 • • • • • • • .. ......... .....• • • • • • 54 • 55 • • • • • • • • • • • • • • • • • • • • 65 vi LIST 0F TABLES Table Page 1 • Results of Group une • • • • • • • • • 2. Results of Group Two • • • • • • • • • 3. Results of Group Three • • • • • 40 Results of Group Four • • • • • • • • • · · · • • • 5. Comparison in Error Rates of The Groups • • • 16 • 0 • 17 • • • 18 • • • 19 • • • 20 6e Group ene, Classification According to Percent Error Rate • 0 • • • • 0 0 • • • 0 • 0 23 • 0 • • 24 o • • • 26 7. Group Two, Classification According to Percent Error Rate • • • • • • • • • • 8. Comparison of Error Rates According to Answer of Question A • • • • • • • • • 9. Comparison of Error Rates According to Answer of Question B • • • • • • • • • ·... 27 ·... 28 • - 10. Comparison of Error Rates According to Answer of Question C • • • • 0 • • • • 11. Comparison of Error Rates According to Answer of Questio'n D • • 0 ··· • • • • • • 29 • • 0 32 12. Group Three, Classification According to Percent Error Rate • 0 • • • • • • • • · 13. Group Four. Classification According to ,-c' Percent Error Rate • • • • • • • • 0 • • • · • 35 14. Percent of Error of Subjects Who Responded to Question "A", and Who were or were not Instructed by the Pharmacist • • • • • · • • • 40 15. Percent of h'rror of Subjects Who Responded to Question "Brr, and Who were or were not Instructed by the Pharmacist • • • • • • • 16. Percent of Error of Subjects V,'ho Responded iii 0 41 to Question "C". and Who were or were not Instructed by the Pharmacist • · • • • • • '" • 42 17. Percent of Error of Subjects Hho Responded to Question "D", and Vlho were or were not Instructed by the Pharmacist • • 18. Comparison of Error Rates by Sex 0 'I • • 19. Comparison of Error Rates by Age • • • viii · • 43 • • · 44 • • • 45 • • · · LIST uF FIGURES Figure Page 1 • Computer Comparison of Group one to Group Two • • · · • • · 55 • • • · 55 • • • • • e • • 56 · . . .. . . . . . . . . . . . . . 57 • • • • • 0 • • • • • 2. Computer Comparison of Group one to Group Three · " • • · • • • • • · • • • 3. Computer Comparison of Group one to Group Four. e " " II · • • . • II • 56 4. Computer Comparison of Group Three to Group Two " .. • tl • • • II • • • · .. • • • 5. Computer Comparison of Group Two to Group Four 6. Computer Comparison of Group Four to Group Three • • • • • • • II • • • 70 Frequency of Errors in Group One 8. Frequency of Errors in Group Two • • • ·.. ·..... ·..... ·.... Frequency of Errors in Group Four • • ·. Questionaire Used at Interview · . . . . . . Pharmacist Check List • • • • • • • · . • • • 9. Frequency of Errors in Group Three 10. 11. 12. 57 58 59 60 61 62 63 13. Sample of Written Instruction for patient • • • ............ . . 64 ABSTRACT This study was to determine the effect of a pharmacists' instruction on the ability of one hundred people to take their medication as their physician prescribed it. These people were patrons of a retail pharmacy, and patients of private physicians. Four groups of twenty five were selected, and each was given a different type of instruction about their medication. first group was given no instruction. was given oral instruction. The The second group The third group was given oral and written instruction. The fourth group was given oral instruction, and had their prescription labeled with a time of day dose schedule. The rate of error was found by determining the ratio of errors to the number of doses the pRtient should have taken on the fifth day after a prescription was filled. The number of errors was determined by visiting the patient at his home, and counting the doses remaining in the prescription containero There was a decrease ~n error from 26% in group one to 16% in group two, and 17% in group three and 1009% in group four. The results of this study indicate a need for instruction to the patient about his medications, and toe effect of four types of instruction be a pharmacist. Chapter 1 INTRODUCTION Today's health care makes use of a variety of potent medications. The effective use of these agents depends on the patient receiving the medication at the correct dosage level, and for the proper duration of time. ~-The need to maintain effective blood levels of medication in present day chemotherapy has fostered many studies to determine the medication compliance of the patient in the home. In a compilation of thirty-three compliance studies, the three methods of measurement used were: pill count, urine test, and patient report. The range of percentage of noncompliance was from six percent to ninety-two percent. 1 In a study to determine how outpatients interpret their prescription directions, the patient was given his prescription, asked to read the label, then tell the pharmacist the hours of the day he planned to take the medication. When the dosage regimen was stated once daily, or at a specific hour, the patients had little trouble deciding on a satisfactory regimen. When the directions of b.i.d, t.i.d, and q.i.d. were used there was a great .. variation in time interval between doses. One patient planned to take his q.i.d. medication at 9 A.M, 10 A.M, 2 11 A.M, and 12 Noon. 2 Edward E. Madden Jr. Chief of Health Services Research Department University School of Public Health and Tropical Medicine Hospital, New Orleans, Louisiana, suggests that great harm to the patient can be the resuIt of incorrect use or lack of use of prescribed medications. He also indicates there is a pointless waste of resources when the physician prescribes a regimen, and the patient does not follow it. Mr. Madden said: The pharmacist has an ideal opportunity to counsel the patient when he dispenses his medication. At present education is an incidental, not an inteeral part of the jobs of most health personel. The educational aspects of health activities are too important to be allowed to occur by chance; planned educational intervention should occur as a result of carefully conceived planning to realize maximal gain from potential educational situations. 3 ~n the chain of events which starts when the patient consults a physician, and ends when the patient either follows or varies from the prescribed regimen, the dispensing pharmacist is the last health professional to infl uence the patient./ 'rJr. Berna.rd G. Keller, Jr. and Raymond J. Bennett classified the pharmacist as the last possible source of error. 4 By enhancing the pharmacists traditional dispensing role to that of drug adviser, many believe be will be able to effect better patient care. There is need to determine clinical significance of deviation from optimal schedule for most drugs. Study 3 on some drugs indicates that variation from a recommended regimen decreases the desired effect of the drugo 2 !' A definition for medication error has been given as: The administration of the wrong medicine or dose of medicine, diagnostic agent or treatment requiring the use of such agent to a patient; or the administration of such agent or treatment at the wrong time, or to the wrong patient; or failure to administer such medication, agent or treatment; or the failure to administer at the time specifieo or in the manner prescribed or normally considered as acceptable practice.5 The study of mtidication errors in a hospital set- ting dates back at least as far as 1937 in a study which indicated a very small amount of error was occuringo 5 It wasn't until the 1950's when sufficient study was made to identify the large amount of error that was taking place. In a study at the University of Florida Teaching Hospital, 572 doses given by nine nurses revealed that one out of six doses was in error. 5 This high ratio of error was found in a hospital setting where the patient had no responsibility for his own medication. This responsibility was that of trained personnel, physicians, pharmacists, and nurses. It is logical to consider the amount of error present when the patient has complete responsibility to take his medication correctly. Medication errors by outpatients have been studied on various types of subjects. In 1963 a study on the failure of children to re- 4 ceive penicillin by mouth indicated that by the third day 44% of the children were taking their penicillin consistently, and by the ninth day there were only 18% doing so. 6 In 1962 a study of medication errors made by aged patients showed 59% of the patients were making some kind of medication error. Part of the source of trouble was voiced by one elderly lady as follows.? The doctor explained it all so carefully, and the nurse went over them too before I left the clinic. He said "Now three of these are medicines that you'll only have to take for a little while, but this one is very important. You are going to have to remember to take this every day or you're going to be in trouble." He even folded the prescription differently for the important medicine. But I don't think he knows that you turn them all in at the pharmacy and wait. Nhen you get them all back, there's no way to know which pills were in the folded prescription, and the heart medicine doesn't look as important as the other kind dose. A study to determine the effect of instruction and labeling on medication errors was reported in 1966. While an error rate of 90% was reported, the instruction and labeling made no difference in rate of errors. 8 A study in 1969 on errors during self administration by thirty men of ages from twenty-one years to ninety years taking a wide variety of drugs indicated errors occurring at a rate of twenty-five percent. 9 In 1972 a study on the evaluation of outpatient counseling about their medications produces the following: In the control group 43.3% of the patients had an error 5 rate over 20% and 30.8% of the counselled group had an error rate over 20%.3 In 1969 a study was reported on an outpatient group of one hundred-eighty indigent patients from a university hospital which sampled various clinic specialties; medical, circulatory, O.B.G.Y.N., surgical, E.E.N.T, and psychiatrice A comparison of variation was made of the groups, and no significant difference was found. The study reported a 42% error. 10 STATEMENT OF THE PROBLEM > There is sufficient evidence in the literature to indicate the presence of a high rate of error when it is the patients responsibility to take his medication as directed. The present health care cycle lacks an individ- ual to provide the education necessary to insure compliance to a dosage regimen. Studies to determine the effective- ness of the pharmacist in supplying this education are in need. It was with this need in mind that this study was undertaken. The purpose of this study was to compare the ratio of error made by patients when four different types of instruction were given by the pharmacist. Ratio of error was determined by dividing the number of errors committedt-by the number of possibilities there were to make errors. 6 The types of instruction the pharmacist is to give the patient group are: 1- No instruction. 2- Oral instruction. 3- Oral plus written instruction. 4- Oral and specific times of day dose regimen instruction. KFurther, it is proposed to determine if physician instructions are a deciding factor in patient compliance to a dosage regimen. The following are hypotheses to be tested. 1- There will be better patient compliance to dosage regimen when the pharmacist orally instructs the patient about prescribed medication than when no instruction is given. 2- i~ben the pharmacist orally instructs the patient about medications, compliance to dosRge regimen will be increased if the pharmacist provides the patient with a written reminder of the instructions. 3- Nhen the pharmacist provides oral instruction about medication, he can further improve patient compliance to dosage regimen if he specifies on the label the time of day each dose is to be taken. 4- Patients who receive oral instruction from their physician about their medication comply better with the dosage regimen for their medicati~n than those who do not receive this instruction. 7 BASIC ASSUMPTIONS Possible factors which may affect patient compliance are age, sex, intelligence, side effects, influence of supervision, forgetfulness, length of treatment, confidence in physician, time spent by physician on instructions about the medication, types of medication, mental status, number or types of diagnosis, and knowledge of purpose of medication. to consider ~hose It is the purpose of this study factors upon which the dispensing pharm- acist may have some influence. They are: effect of label- ing and pharmacist instructions, forgetfulness, confidence in physician, and knowledge of purpose of medication. The basic unit of the study is the number of doses that should have been taken. This has been adopted consider- ing the opportunity for error in hospital medication as follows: 5 1- omission. 2- wrong dose. 3- extra dose given. 4- unordered drug given. 5- wrong dosage form. 6- wrong time for dose given. This study is intended to determine errors in compliance when a patient takes his own medicine at home, and bases the compliance on a one interval pill count. It is not practical to find errors of wrong time, wrong dosage, and unordered drug types. Errors of omission and extra dose were compared with the number of doses the patient should have taken at interview time. 8 DELIMITATIONS The study was limited to the four groups of people selected, and the results are not to be applied to a larger population. It was also limited to a retail pharmacy, and to a routine of four variable sets of instruction as established. 9 FOOTN(;TES Pharma£,I 1Donald E. Franke, Persnectives in Clinical (Hamilton Ill. A.K:WfiitneY~Jr. T977--)pp 44. 2Freya Hermann "The Out Patient Prescription Label as a Source of Medicf4t:ton Errors " American Journal of Hos.E.!!al~rma.£1, pp 155-169 February~-1973-.- - - 3Edward E •. Madden Jr "Evaluation of Outpatient Pharmacy Patient Counselingtl'Journal of the American Eha!ma~eut~£al-h~ociation~ 8 pp 437 A~u;t-1970-.--4Bernard G. Keller Jr. and Raymond J. Bennett, "The Last Possible Source of Error" Journal of the American ~rm~~!ical A~ciati2E.1. NS10 No:13-XligUst197"O:-----5Kenneth N. Barker, Wilson W. Kimbrough, Stud~ .2LMedi~io!! Err.Q!_i.!!~-.!!.Q~pital..:. University mf Arkansas 1966. 6 Abrah~.rn B. Bergman, Richard J. Werner, "Fail ure of Children to Receive Penicillin By Mouth" The New England Journ!l of Medicine 268 No.24 pp 1334-1338. 7Doris Schwartz "Medication Errors Made by Aged P~tients" ~he Amerl:££!n'Jo.EEE~L~!--BE!~i!!B_ Vol 62 NoB pp 51-53 August 1962. SMalahy Bernadine, "The Effect of Instruction and Labeling on· the Number of Medication Errors Made By Patients at Home" Am~ican2,2~~Lof H,2spital Ph~~1..... 23 pp 283-292, June 196b~ 9James C. Clinite and Hugh F. Kabat, "Errors During Self Administration" Journal of The American PharmaceutiEal AS!!.2.2J:gtiQ.!!_9 pp 450-452 September196'9:-------10Clifton J. Latiolais and Charles C. Berry, "Misuse of Prescription Medications by Cutpatients" Drug Intelligence and Clinical Pharmacy, 3: pp 270-277'-Oct. 1969. Chapter 2 DESIGN OF STUDY Criteria The subjects of the study were customers of AlIens Pharmacy, a retail pharmacy in American Fork, Utah. Members of each group were chosen in the following manner. The first three patients of the day having pre- scriptions filled at the pharmacy were selected if their prescriptions fit the criteria for the group that was being established. Patients were selected only on Saturday, Monday and Wednesday, to allow interview days of Wednesday, Friday and Sunday. All prescriptions filled for these patients at this time were discussed at the interview; however only those prescriptions which could be accurately dose counted were used in the study. No individual was interviewed more than once in order to preclude any lasting effect on compliance habit. If any of the first three patients on the above days had been interviewed once, they were omitted and the next prescription patient replaced them. The determination of patient compliance was carried out by an unannounced interview at which time a dOse count was taken and a questionaire was filled out. This inter- view took place five days after the filling of the pre- 11 scriptions. A ~ive day period was chosen to allow the use of short term antibiotic prescriptions. Th~~roup~et the_followin~rit~ria: 1-CONTROL GROUP These prescriptions were filled and labeled as written on the physicians prescription, and no oral instructions (except to answer questions from the patient) were given. If there were questions from the patient about how long to use the drug or what hours of the day they were to be taken, these subjects were deleted from this group. 2-PHARMACISTS INSTRUCTICNS (CRAJJ) GROUP The pharmacist filled and labeled the medication as written on the prescription. Then he instructed the patient orally about the following items. a-What the medication is to be used for. b-The name of the medication. c-How to use the medication. d-'Nhen to use the medication. e-How long to use the medication. f-Any side effects the patient might expect from the medication. g-Wbat to avoid in activities such as driving, working machinery, eating, drinking, and 12 taking other drues. h-Over the counter remedies to avoid. The pharmacist fillea out a check list indicating which of the above items were discussed with the patient. Refer to Figure 12. 3-PHARMACIST INSTRUCTION (ORAL AND WRITTEN) The pharmacist filled and labeled the medication as written on the prescription. Then he instructed the patient orally about the saMe items as listed in group two above. As he orally instructed the patient, he presented to him a printed sheet pertinent to this instruction, and pointed out the applicable statements about this particular prescription. Refer to Figure 13. Before the last group was established, an interview with the physicians writing those prescriptions to be used in this group was held. The purpose of the in- terview was to obtain permission to type time of day regimen instead of the Q.I.D. (four times a day), B.I.D. (two times a day), etc., instructions on the label. Group four had the foll~~~riterl~ 4-SPECIFIC TIME OF DAY REGIMEN. The pharmacist filled and labeled the medication as written on the prescription, except the hour of day the medication was to be taken was typed on the label. The time the doses were 13 to be taken was determined by the pharmacist who considered these things: 1-the type of medication being dispensed, (is it to be taken on an empty or full stomach, or should it be spaced at intervals to include most of the day). 2-the eating and sleeping habits of the patient. The time of meals and time of waking and sleeping was requested before the label was typed. The pharmacist then discused the same items as in section two with the patient. No written in- structions, other than those on the label, were provided to the patient • .Qs2llecti£!!--2E~~alysiL2.Lda.!~ To determine com,liance, the patient was interviewed by one of the two pharmacists involved in the study. The interview was made in the patient's horne, and the patient was not informed of it in advance. During the inter- View, the pharmacist counted the doses that remained in the prescription bottle. To record the information for the study, a questionaire was fill@i $ut. the questionaire is found on page 61. An example of A tabulation of in- formation for each group was made and used to analyze the data. The percent of error was considered to be the .~atio of errors to number of doses that should have been taken at the time of interview. To determine the significance of difference between the percent of error each group had, a computer was used 14 to obtain a "Z" statistic. The "z" statistic is more appropiate for small sample sizes as used in this study, and is used to refer to significance of difference between smaller sample size groups. The" z,r statistic was then compared with a table of normal probability curves. The level of significance was set at the .05 level to determine if the null hypothesis was accepted or rejected. Chapter 3 RESULTS OF THE STUDY When the four groups were established, the routines were set up, and the interviews were made, it was apparent that instruction on compliance was effective, and the pharmacist had influence on the way patients followed a dosage regimen. GROUP DATA The groups have been referred to by number throughout the discussion. Group one was the control, group two had oral instruction, group three had oral plus written instruction, eroup four had oral instruction plus time of day regimen typed on the label. Group One The control group was made up of fifteen female and ten male patients. years to seventy years. The range of age was from twelve This group had opportunities for error of 503, and made 131 errors. This represents an error rate of 26%, and the mean error of 5.2 doses per patient. The mean opportuni ty for error was 20. 1 per,- patient. The results of group one have been tabulated in table 1. Table 1 + +--~---+---t- + - 1-----1 + J.-._-_.--- _··_··-1· . ..1:--1----.-1-------1 t--·'~-1--~~-~--~--1 :F' X X }' _.1U. j_ .. 3-D -9-10 0 22 '2) 24 .25 70 X 65 X _._X 6~ 58 1.6 10. 51 13 F ~.- -.-~ - + + + ..... + + + -- -. + + ....±- --±-. + + GROUP ---- *.- ...-±-+ -----_._- X ~~---- ----- + 1 RESULTS 0F GROUP 1 f = forgot.(+)= answered yes. (-)= answered no + ~ Table 2 / /I! ~;; ~ t ~ ~ t: I tit; :s fr § {J ~!R. ~ §l~~l IJ !} rJ ~ ~· ~'?i:1~~ ~. ~. 4t ~~II 0iftJj&~I'"~if ~~I '., ~~~. t & I"/) ~ F 1 1S 2 11 F ----~ 6 M 3 -·4 - -3cf --.F 5 64 F 6 40 F 7 53 F 8 31 F 54 F 9 10 17 M 11 12 29 11 M M 5 6 F 13 14 15 16 65 17 18 19 20 21 22 X X X X X X X X 71 M X M X 1\1 M M F X X X X X X X X X X X X X X X X X X F I X X X X X X M F X X X X X 5 9 15 23 X --X X X X X 66 7 24 -25 --- F F M 42 23 11 23 x X X X X 15 9 0 2~__ _0 31 3 14 4 20 3 1 34 12 0 24 3 .20 3 20 2 11 1 16 3 21 3 24 1 1 11 2 10 0 29 10 28 5 -- 21 8 36 0 20 11 14 15 12 0 0 0 77 ~88 GROUP 3 = forgot. (+) = f f - -----±-- -+ ,..".- - + + - -+ -+ + + + + - - + + _T __ ------ - + - -+- +.+ + + + + + + + + f - -+ + -+ + - - - - + + - - - - + + + + + + + - + + + - + + + + ~ -+ + - + + + 2 RESULTS GF GROUP f f 2 answered yes. (-) answered no •. .- 3 Table ·1;f;&I </Ij i(li i iJ ~ § ~ Q q ~I f/ . f/ !.j 0;) l; ,..., & tll§ I/!; ; - iI:;>~: rfl0~~g, £j iI (:j 0 "'l ~~ 'fi 2 26' .:1 ; 20 X M X X 10 20 i 15 b X 24 5 F X X 21 0 5 5 F F 4. . 40 5 54 -E 60 13 10 15 130 123 6 7 8 9 10 11 ~23 '25 1 :29 ,-15 _ 16 16 9 17 2, 58 18 19 53 20 12 21 16 22 38 48 46 47 23 24 25 ·X M X X X ·X X 0 14 0 0 M F X X 18 X X F X X X 25 15 12 20 17 20 17 8 8 M F X X F X X M F F M X X X X F X X F F F F M X x X X X M X F M X X I X X X X X X X X x X X X X a 1 0 : + + + 1~ 12 10 TO 24 12 21 24 12 12 12 21 12 ~14 71 GROUP 3 -- ~ - + + + + - + + + + + + + + - -+ - f!/ ~ ~~ :$,§' + + + + + + + f"- - + + + --=- - 1- -+ - --- + - - + - - + + - + - - + - + + + + 17 12 I -t- + 17 ~ ~- ~ t::J• • ~ ~- ~. ~t::; ~c: I 1 115 ~ & - -~---- !I RESULTS CF GR0UP 3 f = forgot_ (+) = answered yes. (-) = answered no. K Table 1 2 X X x X x x 3 4 5 :B' X + + x + + + X X X X X X ..-X X + -+ x x X M + X X X rl 4 - X x x X X X + F + F + - + F + + + - X - X X - X F + - X F + - - + + + 46 GROUP F = forgot. (+) = 4 RESULTS OF GRC UP 4 answered yes. (-) = answered no. Table 5 C(;MPARIS0N IN ERROR RATES OF THE GROUPS 26°0 % 25 24,- 26 t- 23,22'- - 21 20,- 19I- 17.4% 18i- PERCENT GF 17 - 15.7% 16,- 15 14,f ... roSES TAKEN IN ERROR 13,... 12 ... 1 11- 10.9% ' 10,. 9 - 8~- - 7 6,- - 5 3 21o 4 .... j GROUP 1 GROUP 2 GR0UP 3 GROUP 4 21 Qroup~Tw£ Group two was made up of fourteen female and eleven male patients. The range of age was from five years to seventy-one years. they made 77 errors. The opportunity for error was 488, and The error rate was mean error was 3.08 doses per patient. 15.7%, and the The mean oppor- tunity for error was 19.5 doses per patient. The results of group two have been tabulated in Table 2. In group three, there were sixteen female and nine male patients. sixty years. The range of age was froM nine years to The opportunity for error was 414, and they made 71 errors. This was an error rate of 17.4%, and the mean error was 2.8 doses per patient. The mean opportunity 16.5 doses per patient. The results of group for error was three have been tabulated in Table 3. Group :Four ---------Group four had fourteen female and eleven male members. years. The range of age was six years to eighty- one This group had opportunity for error of 422, and made 46 errors. This was an error rate of 10.9%, and the mean error was 1.8 doses per patient. The mean opportunity for error was 16.9 doses per patient. The results ofrgroup four have been tabUlated in Table 4 •. 22 COMPARISON OF THE GRuUPS In testing the hypotheses, each group was compared with the control group. Of interest so was a comparison of each group to the others. ~~~i80E~!_Qroup Two to GrouE-On~ A comparison of Group Two with Group One showed Group Two to have fewer errors, fewer possibilities for error, a similar range of age, and a smaller mean of errors per patient. Group Two had a 16% rate of error while Group One had a 26% rate of error. rates. Table 5 compares group error The difference in rate of error of these two groups was significant to the .01 level as determined by the liZ" statistic. The computer printout of the "ZIf statistic is found in Figure 1 of the appendix. Tables 6 and 7 compare the percent of the people that committed the various percents of error from 0% to 100%. In listing the data cumulatively, it showed that 16% of the people in Group One had over 50% error, while only 8% of Group Two had over 50% error. The sacie tables indicate that 32% of Group One had over 25% error, while only 20% of Group Two had over 25~ error. In order to compare the data about the physiCian instructions shown on Tables 8 to 11, the questions have been referred to by the letters A. B. C. and D. "A" in- dicating the question; "Did your doctor tell you how often ·23 Table 6 # PEUPLE % PEOPLE 96-100 0 0 100 91-95 0 0 1()() 86-90 1 4 100 81-85 0 0 96 76-80 0 0 96 71-15 0 0 96 66-10 1 4 96 61-65 1 4 92 56-60 0 0 88 51-55 1 4 88 46-50 1 4 84 41-45 0 0 80 36-40 1 4 80 31-35 2 8 16 % ERROR CUM % ------------- -------------- _.-_------ ..... ----- ------------- 26-30 -----------_ ... --- ..... 68 1 4 ----.----.. ----_ ... ------- ... -- ------------- 21-25 2 8 64 16-20 0 0 56 11-15 4 1 56 6-10 1 4 40 1-5 2 8 36 1 28 28 0 TuTAL -100 --25 GROUP 1 CLASSIFICATION ACC(;RDING TO PERCENT ERRCJR RATE .- 24 Table % ERROR - # PIDPLE 7 % PIDP1E 100 0 0 100 95 0 0 100 86 - 90 0 0 100 - 85 0 0 100 76 - 80 0 0 100 - 75 0 0 100 66 - 70 0 0 100 - 65 0 0 100 56 - 60 1 4 100 - 1 4 96 96 91 81 71 61 51 55 .. --------------------------------------0 0 46 - 50 ---~-------- 92 41 - 45 1 4 92 36 - 40 1 4 88 35 1 4 84 31 % CUMULATIVE I I 4 -------------------------- -------------_. ~-----------0 0 76 21 - 25 12 16 - 20 76 3 16 1 1 - 15 64 4 26 6 1 - 0 TOTAL 80 30 1 10 3 12 48 5 1 4 36 8 32 32 25 100 .- GROUP 2 CLASSIFICATIGN ACCORDING TO PERCENT ERROR RATE 25 to take the medication?" "B" indicating the question; "Did "c" your doctor tell you the purpose of the medication?rr indicating the question; "Did your doctor tell you how long to take the medication?" liD" indicating the question; "Did your doctor tell you when (What hours of the day) to take the medication?". While these questions ask, "did your doctor tell you ••• ", They really indicated what the patient remembered of the physicians instructions, and not the actual instruction. The following information about the answers to questions A, B, 0, and D by those in groups one and two has been illustrated in tables 8 thru 11. The patients of Group One that answered ""'-ves to ques~ion "A" had a 23% error rate, while Group Two had an 11% error rate. Those of Group One that answered no to question "A" had a 31% error rate, while Group Two had a 20% error rate. The patients of Group One that answered ~~~ to question "B" had a 23% error rate while Group Two had a 14% error rate. Those o·f Gro1:1p One that answered !!.£ to question "B" had a 35% error rate, while Group Two had a 13% error rate. The patients of Group One that answered ~~ to question "C" had a 12% error rate while Group Two had'-a 13% error rate. Those of Group One that answered no to question "C" had a 29% error rate, while Group Two had an 18% error rate. 26 Table 8 COMPARISON OF ERROR RATES ACCORDING TO ANSWER OF QUESTION A -. - QU ESTION:Did your doctor tell you how often to take the medication? Group 1 Group 2 numbel~ 16 13 Group 3 Group 4 3 10 Total 42 YE S % error 23% 1% 25% 19% - number 21 15 10 55 NO % error 20% 31% 12% 9% 14% -~---.- 2 FO ROOT 1 3 25 100 --- TO TAL 25 25 25 - - 27 Table 9 COMPARISGN OF ERROR RATES ACCORDING TO ANSWER OF QUESTICN B ----_ - - -----_._--- .... -- QUESTION:Did your doctor tell you the purpose of the medication? Group 2 Group 3 I Group 4 Group 1 tal j 17 21 number 15 12 19 ! I I YES % error 23% 21% 14% I 13% ! 19% I ----- - number NO 5 9 4 25 1 I --_. % error FORGOT TOTAL 13% 35% 1 25 f/I 1 25 6% 1% 1 I I 14% 3 -- 25 25 100 28 Table 10 COMPARIS0N OF ERROR RATES ACCORDING TO ANSWER OF QUESTIuN C "QUEST IC N:DIdyour'-ifo'ctor'-tel-i--'y'ouho Vi-long ---to-fake -fhemedication? ~ 1 Grou~_G_r_O_U_p__2~~G_r_o_U__P__3~~G_r_O_U_p__4___+-T___O_t__a__l. _____ number YES I i 8 15 15 46 8 ~-----+--------~----- :rror I 12% 16% 13% 6% 14% I--·-----~---~----·----~---------~-------+---------~------·------- number NO 16 % TOTAL 51 --~------~--------+---------+---------- error FORGOT 16 10 9 29% 18% 1 1 25 25 18% 25 13% 21 1 3 25 100 29 Table 11 COMPARISON OF ERROR RATES ACCORDING TC ANSWER OF QUES.J..ION D .. - r---- ~------- ---- QUESTI(;N: Did your doctor tell you when (what hours of the day) to take the medication? Group 2 I Group 3 lP 6 YES Group 4 , Total 16 0 3 : % ! I error 10?( 22% 0% 11% 17% i 1 number 18 17 22 81 24 i j ! INO '- % error 15%,i 28% 18% 11% 17% i .-.-~- FORGOT 1 1 I I ------ TOTAL -- : 1 I 3 , 25 25 ! 25 ,25 100 30 The patients of Group One that answered question "D" had a 10% error rate. 22~b ~~ to error rate, while Group Two had a Those of Group One that answered no to question "D" had a 28% error rate, while Group Two had a 15% error rate. The male patients of Group Cne had a 20% error rate, while those of Group Two had an 11% error rate. The female patients of Group ene had a 31% error rate, and those of Group Two had a 19% error rate. ~~rison of Group Three to Gr~~ On~ A comparison of Group Three with Group One showed Group Three to have fewer errors, fewer opportunities for error, and a smaller mean error. The age range was from nine years to sixty years in Group Three, and from twelve to seventy in Group One. Group One had four people under twenty years, while Group Three had eight. Group One had twelve people between twenty years and fifty years, while Group Three had thirteen. Group One had nine people over fifty years, while Group Three had four. been tabulated in Table 3. This data has Group One had more people in the older age bracket, and also an older age range. Group Three had a 17.4% rate of error. The diff- erence in rate of error between Group Three and Group One .C was 9 percentage points. This was significant to the .01 level as indicated from the computer. "z" statistic obtained by The computer print out is found in figure 2. 31 Comparison of cumulative percent of people committine error indicates 16% of Group One had over 50% error, while 8% of Group Three haa over 50% error. tabulated in Tables 6 and 12. The data has been These two Tables also in- dicate 32% of Group One had over 25% error while 12% of Group Three had over 25% error. The patients of Group One that answered ~ to question "A" had a 23% error rate, while Group Three also had a 23% error rate. Those of Group One that answered no to question "All had a 31% error rate, while Group Three had a 12% error rate. The patients of Group Three that answered question tlB" had a One. 21~fo Y..!l..§. to error rate compared to 23% in Group Those of Group Three that answered nQ to question "B" had a 1~h error rate compared to 35~0 in Group One. The patients of Group Three that answered Y.!!.§. to question "C" had a 16% error rate cOMpared to 12% in Group One. Those of Group Three that answered ~ to question "c" had an 18510 error rate compared to 29% in Group One. The patients of Group ~rhree that answered ~ to question "D" had an 11% error rate compared to 22% in Group One. Those of Group Three that answerea no to Ques- tion liD" had an 18% error rate compared to 28~f in Group One. The male patients of Group Three had a 10% error rate compared to a 20% error rate in Group One. Female patients of Group Three had a 21% error rate compared to 32 Table 12 % ERROR # % PIDPLE PIDPLE CUMULATIVE % 0 0 100 95 0 0 100 86 - 90 0 0 - 85 1 4 100 76 - 80 0 0 96 - 75 0 0 66 - 70 1 4 - 65 0 0 92 56· - 60 0 0 92 96 - 100 - 91 81 71 61 - .... --- ... --51 , . 100 96 96 . 92 0 0 _- ........ -------------_. ------------------------0 55 ... 92 46 - 50 0 - 45 0 0 92 36 - 40 1 4 92 - 0 0 88 4 88 8 84 41 31 35 1 26 - 30 -- ........... ---_ ..... _-- ~--------~2 21 25 - ..... -- --------------- 1------------ 16 - 20 16 16 76 11 - 15 1 4 60 10 4 16 56 "5 1 4 40 9 36 36 25 100 6 1 0 TOTAl i GROUP I 3 CLASSIFICATION ACCCRDING TO PERCENT ERROR RATE .- 33 31% in Group une. 2.2.!!!~ris.2.E_.Qf sr.!:oup Three to Grou~.Q. A comparison of Group Two with Group Three showed Group Two to have more errors and more opportunities for error. The range of age was similar, however the range of Group Two was wider with more people under twenty years of age, and more people over fifty years of age. Group Three had a slightly higher rate of error than Group Two. Considering the number of errors and the number of opportunities for error, this difference in percent was not significant and could be attributed to chance. 2.2.!!!.Ear i ~.!!_ 0 f .J!~~ Fo Er t~ Gr.2E.E--2 n e A comparison of Group Four to Group One showed Group Four to have fewer errors, fewer opportunities for error, and a smaller mean error. The age range of Group Four was similar to that of Group Gne, however Group Cne had fewer people under twenty years of and age, and more people over fifty years of age. Group Four had a 10.9% error rate. The difference in rate of error between Group Four and Group One was 15.1 percentage points. This was significant to the .01 level, as indicated from the "Z" statistic obtained by computer. The print-out is shown in Figure 3. Comparison of cu'rnul- ative percent of people committing error indicated 16% of Group Cne had over 50% error, while 0% of Group Four had over 50% errorG The data has been tabulated in Tables 6 34 and 13. The same Tables indicate 32% of Group One had an error rate over 25%, and 8% of Group Four had an error rate over 2 55b. The patients of Group Four that answered ~ to que s t ion "A" ha d a 1 9~~ err 0 r rat e com par edt 0 2310 in Gr 0 up One. Those of Group Four that answered ~ to question "A" had a 14% error rate compared to 31% in Group One. The patients of Group Four that answered question "B" had a One. 13~b to error rate compared to 23% in Group Those of Group Four that answered no to question JIB" had a 6% error rate compared to 35% in Group One. The patients of Group Four that answered question "Gil had a Cine. ~~ 6~b error rate cOMpared to 127~ ~~ to in Group Those of Group Four that answered no to question "c" had a 13% error rate compared to 29% in Group One. There were no patients in Group Four that answered ~~ to question "D". The male patients of Group Four had a 15% error rate compared to 20% in Group One. Group Four had an 8% Female patients of error rate compared to 31% in Group One. Group Four had fewer opportunities for error, and fewer errors than either Group Two or Group Three. The range of age is similar in Groups Two, Three, and Four. The difference in percent of error between Group 35 Table 13 - .. .... ,--"-------- -.-.. ~ ' % ERROR 96 - 911 -------~------ # PEOPLE % PIDPLE CUMULAT IVE %I 0 0 100 - 95 0 0 100 86 - 90 0 0 100 - 85 0 0 100 76 - 80 0 0 100 66 - 70 0 0 100 61 - 65 0 0 100 56 - 60 0 0 100 51 - 55 0 0 100 81 100 -~---------- --------------------------------------0 0 100 46 - 50 41 - 45 1 4 100 36 - 40 0 0 96 31 - 1 4 96 35 26 - 30 2 8 92 -------------.---------~-------------- -----------21 - 25 12 3 84 ..... 16 - 20 3 12 72 11 - 15 3 12 60 6 - 10 4 16 48 1 - 5 0 0 32 8 32 32 25 100 0 TOTAL .. GROUP 4 CLASSIFICATION ACCORDING TO PERCENT ERR0R RATE - 36 Four and either Two or Three was significant to the .01 level. Chapter 4 DISCUSSICN OF THE FINDINGS The results of the study revealed that the pharmacist could greatly influence patient compliance to dosage regimen, if he structured his activities to that end. The control group had their prescriptions filled in a manner very similar to that being done in most retail pharmacies today. When the various types of instruction were used, the error rate diminished from 26% to 16%, 17% and finally 11%. The fact that patients do need assist- ance in understanding their medication was consistent with the literature referred to in Chapter One. __ /V By filling in this void in the heal th care chain the pharmacist would have the opportunity to enhance his professional image, increase his service in the health field, have increase in his immediate business endeavor, and increase the long range demand for his services. ANALYSIS OF THE PROBLEMS ~Compliance of those who received oral instruction from the pharmacist was greater than those who did not. Analysis of the data indicates that null hypothesis Number one should be re jected. ,/The difference in patient compliance, expressed by percent of error, was significant to the .01 level. 38 While it was anticipated that the use of a written reminder in addition to oral instruction would increase patient compliance, the results of tbis study did not support this. Compliance of those who had written re- minders and oral instruction from the pharmacist was less than those who had only oral instruction. The difference in compliance was not significant to the .05 level, therefore it had to be attributed to chance. It was apparent from the home interviews that the patients did not read the written reminder. The~ listened to the pharmacist, watched him point to the printed section on the paper, and thought they understood the instructions so they discard- ed the paper. 'Since the difference between Group Two and Group Three was insignificant, Group Three became essentially another oral instruction group. Vfuile this com- parison waS not stated as part of the study, it did support the effectiveness of oral instruction by the pharmacist. ,The greatest difference in compliance of the groups was that between those who had no instruction (Group ene) and those who had oral instruction plus dosage regimen (Group Four) typed on the prescription label. The diff- erence was significant to the .01 level, so the null form of hypothesis Number Three had to be rejected. ,- The dis- cussion prior to labeling with dosage regimen had to be considered part of the reason for the improvement in compliance. 39 The patients were led by the pharmacist to determine the best time for them to take the medication. The label then served as a reminder of the time decided upon. This group provided the greatest challenge to the pharmacist when the initial contact was made with the patient. In dis- cussing living habits with the patient and arranging a time for the various doses to be taken, it was easy to see why many have poor medication compliance. ~~en one con- sidered time of waking, time of meals, time of retiring, and working habits, it was sometimes difficult to arrange an optimum dose time for medication. This waS true par- ticularly when medication was intended to be used with, or without food, and when there was an optimum time interval consistent with the metabolism of the drug. How- ever when this waS decided and printed on the prescription label, it was a helpful guide for the patient in remembering each dose. Many of the people interviewed mention- ed how this helped them fix a time in their mind. Better compliance was achieved when both the physician and the pharmacist instructed the patient about their medication. Also lower compliance was achieved when neither physician nor pharmacist instructed the patient. This was supported by the data on all of the questions answered during the interview except the one asking, "dId your doctor tell you the purpose of the medication." Of the one hundred people in the study, twentynine had both physician and pharmacist instruction about 40 how long to take the Medication (Question A). of error waS 19~. Their rate Ten people had neither physician nor pharmacist instruction, and their rate of error was 31%. The data is illustrated in Table 14 below. Table 14 RECEIVED PHARMACIST INSTRUCTILN YES NO Groups Group 2, 3, 4 1 number of 29 13 . people I YES RECEIVED percent of error PHYSICIAN number of people percent of error INSTRUCTION NO 19% 23% 10 45 13% 31% PERCENT OF ERROR OF SUBJECTS WHO RESPCNDED TO QUESTION "AU (HOW LONG ARE YOU TO TAKE THE MEDICATION), AND WHO WERE OR WERE NOT INSTRUCTED BY THE PHARMACIST 41 Fifty three people had both physician and pharmReist instruction about the purpose of the medication (Question B), and their error rate was 16%. Five people bad neither physician nor pharmacist instruction, and their error rate was 6%. Tbe data is illustrated in Table 15. Table 15 RECEIVED PHARMACIST INSTRUCTION YES NO Gr ou p S 2, 3, 4 YES number of people RECEIVED percent of error PHYSICIAN number of people INSTRUCTION NO Percent of error 53 Group 1 19 16% 5 23% 5 35% PERCENT OF ERROR OF SUBJECTS WHO RESPONDED TO QUESTION "B" (WHAT IS THE PURPOSE CF THE MEDICATION), AND WHO WERE OR WERE NOT INSTRUCTED BY THE PHARMACIST 6% 42 Thirty-eight people had both physician and pharacist instruction about how long they should continue the medication (Question C), and their error rate was 11%. Sixteen people had neither physician nor pharmacist instruction and their error rate was 29%. The data has been illustrated in Table 16. Table 16 RECEIVED PHARMACIST INSTRUCTION NO YES Group Groups 1 2, 3, 4 number of 8 38 people YES RECEIVED PHYSICIAN INSTRUCTION percent of error number of people NO percent of error 12% 11% 16 35 16% 29% PERCENT OF ERROR OF SUBJECTS WHO RESPONDED TO QUESTION "C" (HOW LCJNG SKOULD Y(;U TAKE THE MEDICATION), AND WHO WERE OR WERE NOT INSTRUCTED BY THE PHARMACIST 43 Nine people had both physiciRn and pharmacist instruction about when to take the medication (Question D), and their error rate was 7%. Seventeen people had neither physician nor pharMacist instruction and their error rate was 28~. The data has been illustrated in Table 17. Table 17 RECEIVED PHARMACIST INSTRUCTICN YES Groups . ·2. number of people YES RECEIVED PHYSICIAN INSTRUCTION NO 4 Group 1 7 9 percent of error number of people percent of error 3----, NO 7% 22% 17 64 15% 28% PERCENT OF ERROR OF SUBJECTS WHO RESPCNDED TO QUESTION IfD" (WHEN SHOULD YOU TAKE THE MEDICATION), AND WHO WERE OR WERE NOT INSTRUCTED BY THE PHARMACIST Those with no pharmacist instruction had the highest rate of error regardless of how they answered the question. An interesting deviation from the anticipated was the result of medication compliance differences between males and females. The belief that the females would 44 Table 18 COM?ARISGN OF ERROR RATES BY SEX Group number 10 I MALE % error number FEMALE Group 2 IP 11 Group 4 Total 11 41 - 20% 15 11% 10% 14 15% 16 14 14% 59 -- % error 19% : 21% 8% 20% I TOTAL 25 25 ~ 100 45 Table 19 COMPARISCN 0F ERRCR RATES BY AGE --- ------Jnumber ! Group 1 - Group 2 Group 3 12 4 Group 4 8 Total ~--- .--~-. - ---- 31 7 ,I UNDER 20 YEARS I % --- 25% I ! error 18% 11% i 1 ! number I 12 7 12 13 4 i i 20 YRS TO 50 YRS - "--~"-t---~-- i I j % 36% error 25% 21 13% .- -.--.-~ number 9 6 6 4 OVER '50 YRS % error 17% TOTAL 25 17% 25 9% 25 1 5 .i.- ~-~"~----,---~ .. 46 be better able to remember to take medication was opposed by the results for the total of one hundred people, and for all of the groups except Group Four. The data is illustrated in Table 18. The age group with the highest error rate was the twenty to fifty year old group. It was anticipated that the elderly would have the poorest compliance, however, in this study they had the same rate as the under twenty years, which was better than the twenty to fifty year old group. They seemed to have more time to remember their medications, and more people to help them remember. RELATIGNS WITH PATRONS CF THE PHARMACY The initial contact with the patient occured when he or she presented a prescription at the pharmacy. prescription was filled and sold to the patient. This The discussion which took place at this point revealed that the layman was knowledgeable about medications, and disease states in general, and that he was interested in knowing more about the specific medication prescribed and how it would affect his problem. There were many incidents when the patient mentioned his appreciation of the pharmacist for taking time to help him with his medication problems. Simple suggestions such as "you might take wi th orange juice", "take wi th out milk", "this may make you drowsy". or "take on an empty stomach" were appreciated. Sometimes there were statements 47 which pointed out past needs for instruction. One lady, when advised not to take metronidazole and drink alcoholic beverages at the same time said, "Now someone tells me. I had this same medication a few months ago when away from home, had a few drinks with friends, and became ill". This particular lady was not in the study, however, the instruction given was the same as for those participating. Only those chosen according to plan were used in the study, however, all patrons except the control group were routinely given the oral instructions, because of the patient response. Contact at liome Interview There was no one who refused to be interviewed, and nearly all were very receptive and anxious to discuss their medications. After the information for the study waS obtained, often the patient desired information about other medications they had taken or medications others in the family were taking. Most were interested in the study, and willing to participate. One individual commented she previously had a pattern of poor medication compliance, but when the time of day regimen was on the label she found it conducive to taking her medication. She had no errors in dosaee during the study. There was open discussion in most homes, and all appeared to give honest reasons for making errors, or indicated they did not know why the error was made. 48 Many indicated they used the pharmacist as a source of information about medications. Bepe~tC£nt~ts Those of the study who had other prescriptions filled after they had been interviewed expected to receive instruction about the new medication, and were quite willing to take the advice given. These people were also more interested in obtaining information about over-the-counter (O.T.C.) medications. PHYSICIAN RESPONSE All of the physicians interviewed were aware of the problem of poor medication compliance, were willing to become part of the study, and were interested in the results. The physicians are all on the staff at the same hospital, and they discussed the study at a staff meeting giving it their approval. Special consideration was given in situations when the physician desired the pharmacist instruction to be different from the routine established by the pharmacist. Effort was made to compliment the instructions given by each physician to his patient concerning individual medications. When interviews for the four groups were completed, the medical staff secretary asked for a copy of the results. A report was prepared, and he presented it at a staff 49 meeting. STRUCTURAL ELEMENTS OF PHARMACY R(;UTINE The pharmacist is strategically located in the health care cycle to give instruction on medications. To provide this service he must organize his sources of information about the medications he dispenses to have its ready access when he fills a prescription. To implement a program of instruction for the purpose of this study, a file of instructions for the generic drugs was prepared, and a number coresponding to a page of this file was attached to the shelf below each bottle containing this medication. When the pharmacist filled a prescription for a drug he consulted the instructions applicable, and discussed them with the patient. The instructions for each drug were general, and had to be applied to the individual patient, considering other medications, patients life pattern, and physician instructions. EFFECT OF STUDY uN PHARMACY BUSINESS The gathering of data for the study took place over the time span of one year. Shortly before the begin- ing of the study, a large chain drug store opened in a .(- city close by. Competition for prescription business was more intense than it had been previously. During the year the pharmacy experienced an increase which was double that of the rest of the store. The main variation from past 50 methods of doing business was the inclusion of patient instruction about medications. Chapter 5 SUMMARY THE PROBLEM During the 1960's there were many studies on the amount of error made by patients in taking their own medication. The patients involved in these studies were hospital outpatients. Since a high rate of error was in- dicated by these studies, others have been undertaken to find methods to improve patient compliance. Much has been written about the practicality of having pharmacists provide education to patients about their medication. Since the pharmacist is the last pro- fessional,contacted by the patient, in the health team cycle, many feel he could be effective in instructing the patient, thereby decreasing medication errors. METHOD The people in this study were patients of private physicians, and had prescriptions filled at a retail pharmacy. The pharmacy was Allens Drug and Food Fair, located in American Fork, Utah. K A total of one hundred people were involved in the study, and they were divided into four groups. The first group received no instruction from the pharmacist, and 52 served as a control. The second group received oral in- struction about their medication. The instruction was to help tbem understand; 1-The purpose of the medication 2-The length of time they should take the medication 3-Any side reactions tbat might develop 4-The importance of taking tbe medication at the correct time of day 5-Vfuat to avoid ingesting while taking the medication. The third group received oral instruction and a written reminder of the instruction. The fourth group received oral instruction and had a time of day dosage regimen typed on the prescription label. Five days later a pharmacist visited with the patient in his home, interviewed him, counted the remaining doses of medication, and filled out a questionaire. There was no advance notice given about the interview. FINDINGS From the doses remaining, the number of doses taken in error was determined. A ratio of doses taken in error to doses that should have been taken provided the percent of patient error. Group One took 26% of their doses in error. Group Two took 15.7% of their doses in error. Group Three had a 17.4% error rate. Group Four bad a 10.9% error rate •. ..C0NCLUSIONS The results of the study indicated the pharmacist 53 could affect the compliance of patients to a medication schedule.~Also the type of instruction the pharmacist provided did cause the patients compliance to vary. The written reminder was not effective in improving compliance. Oral instruction by the pharmacist did help the patient comply wi th the directions of the physician. Vv'hen dis- cussion of the patients living habits led to an optimum dosage time, the lowest rate of error was achieved. To become an effective factor in improving patient compliance, the pharmacist must train himself to consider and understand the problems the patient has when taking medication. SELECTED BIBLIOGRAPHY 1. Donald E. Franke. per~ecti ves in Cl.!,nical (Hamilton Ill. A.K. W 1tney Jr. 19~ )7 Ph~Ema.2y. 2. Freya Hermann. "The Out Patient Prescription Label as a Source of Medication Errors." American Journal of Ho spi tal Pharmacl.!. February, 1973:----------3. Edward E. Madden Jr. "Evaluation of Outpatient Pharmacy Patient Counselingll Journal of the American Pharmaceutical Assocrauan--g--:-AuBus1;, 19'73. 40 Bernard G. Keller Jr. and Raymond J. Bennett. "The Last Possible Source of Errorll Journal of the American Pharmaceutical Association NS10 No .---s-:Augus""t 19 /0 1 5. Kenneth N. Barker, Wilson W. Kimbrough. Study of Medication Error in a Hospital. University of Arkansas. 1966. 6 0 Abraham B. Bergman, Richard J. Werner. "Failure of Children to Receive Penicillin By Mouth" The New Englan£_![our~l oLMe£ic..!~. 268 NG. 24. ---..- 7" Doris Schwartz. "Medication Errors Made by Aged Patients" 1.Ee Americ~E_Jo.!!!'nal--Ei2~~in~. Vol 62 No 8 August, 1962. 8. Malahy Bernadine "The Effect of Instruction Rnd Labeling on the Number of Medication Errors Made By Patients at Home" American Journal of Hospital ~har~2.2~ • 23 June',"-1966. ----- 9. James C. Clinite and Hugh F. Kabat "Errors During Self Administration" Journal of The American Pharmaceutical !ssociatioE 9 Sept~;;-,1 9 6 9 : - - - - - - - - - - - - - 10. Clifton J. Latiolais and Charles C. Berry "Misuse of Prescription Medications by Outpatients" Dru.,g Intell..!.gence and Clinical Pharmacy 3 Oc.to ber, 1969. IX :B'IG cc CC Filj? 131 503 77 ASS 0.26043 O.1577B I 1 C}' GHC UP TC s ... . ~~rror in Group 1 S •• o.(pportunities for error in up 1 S .••. Brror in Group 2 S •••• Lpportunities C :" • ••• /'.i • ••• I error in Group? Error in Group 1 Brror in up? 3.99571 .Al. •••• z statistic Ji' lG UH:~~ 2 CU MPUTB~1 tVIPAHISCN GRCUP C. 503 S •••• Error in Group 1 S •••• cpportunities r error in Group 1 71 414 S •..• ~rror in Group 3 S ••.• Lpportunities for error in Group 3 131 in Group 1 in Group 3 0.26043 AO •••• 'Jr, 0.17149 BO •••• ~rror 3.24235 statistic • ••. l ~rror 56 FIGURE 3 CGMPUTER CUMPARISCN CF GReU? CN}J TC GRCUP FOUR 131 503 So ••• Errors in Group 1 S •••• Gpportunities for error in Group 1 46 422 S •••• Errors in Group 4 S •••• 0pportunities for error in Group 4 0.26043 O. 10900 5.85123 AO •••• % Error in Group 1 .P() •••• % Error in Group 4- AC •••• Z statistic FIGURE 4 CUMPUTER COM?ARISGN GF GRC UP THREE TO GRCUP TWC; 77 488 S •••• Errors in Group 2 S •••• Gpportunities for error in Group 2 71 414 S •••• Errors in Group 3 S •••• upportunities for error in Group 3 0.15778 0.17149 -0.55528 AU •••• % Error in Group 2 in Group 3 BC •••• % Error AO •••• Z statistic 57 FIGURE 5 CCMPUTER CGMPARIS(jN OF GRCUP T'NG TO GROUP FOUR , ... 77 488 in Group 2 S.o •• Cpportunities for error in Group 2 46 422 S •••• Errors in Group 4 S •••• upportunities for error in Group 4 -:"\ b •••• l!irrors 0.15778 AL •••• 7~ Error in Group 2 0 10900 .Be •••• % Error in Group 4 0 2.16031 AO ••• Z statistic 0 FIGURE 6 COMPUTER COMPARISON OF GRGUP FCUR TO GROUP THREE 414 S •••• Errors in Group 3 So ••• Cpportunities for error in Group 3 46 422 S.o •• Cpportunities 71 S •••• Errors in Group 4 for error in Group 4 0.17149 AO •••• % Error in Group 3 ]3(j •••• % Error in Group 4 o. 10900 2.61793 AO •• 0 • Z statistic FIGURE FREQUENCY OF ERRORS IN GROUP 1 22 - 1 - - - - - + 21 - 20 - 19 18 - 17 NUMBER OF EHRCRS 14 - 13 - 1 - - - - - - 1 12 11 10 9 8 ·7 6 5 4 3 2 1 o 1 2 3 4"5 67 8 NUMBER OF PEUPLE MAKING ERRORS 9 FIGURE FREQUENCJ uF ERRORS IN gROU:p NUMBER OF ERRORS 22 21 20 19 18 17 16 15 14 13 12 .. .. - -.. - .. - - 10 .. 9 8 - I 11 7 - 6 .. 5 4 3 - I I J - J 1 - 0 - ~ 2 2 ~- .-(- I 1 I 2 3 4 NUMBER OF PEOPLE , 5 I I 67 MAKING ERRORS I 8 9 FIGURE FREQUENCY OF ERRORS IN GRGUP ,22 21 20 19 18 17 NUMBER 16 15 14 OF 13 ERRORS 12 11 10 9 8 7 6 5 4 3 2 1 0 3 \ •• - - - - - I - - -_J - - - I - - I I I .«- I 1 2 :3 4 5 6 7 NUMBER OF PEOPLE MAKING ERRORS 8 9 FIGURE FREQUENCY OF ERRGRS IN GR0UP 22 21 - 4 .. 18 17 16 15 14 13 12 11 10 9 8 7 6 5 - 20 19 NUMBER O}4"' ERR(;RS 4 3 2 1 O· - I - - I - J I 1 2 3 4 5 , .'(- I 6 7 NUMBER OF PEOPLE MAKING ERRORS 8 9 FIGURE QUESTICNAIRE USED AT INTERVIEW NAME_ _ _ _ AGE SEX_MARRIED_SINGLE LIVING WITH SOMEONE__ ALONE__ REASON MEDICATION WAS PRESCRIBED_____ __ NAME OF MED. ____-PRESCRIPTION NUMBER_____ DDSES PRESCRIBED PER DAY___ DUSES DISP. noSES REMAINING AT INTERVIEW__ __ roSES USED___ roSES THAT SHOUI.D HAVE BEEN USED)_ ERRORS___ DATE OF INTERVIEW 1-IF roSES WERE SKIPPED, WHY? OR EXTRA DOSES TAKEN, WHY? A-Patient misunderstood instruction____ B-Patient forgot doses___ C-Patient thought drug stopped working___ D-Patient thought he was cured___ E-Patient disliked medication F-Patient lost medication___ G-Patient got medications mixed up___ H-Patient thought he needed an extra dose I-Medication made the patient i l l _ J-Other -- 2-DID YOUR DGCTOR? A-Tell you how often to take the medication? What were his instructions? B-Tell you the purpose of the-medicatIon? What were his instructions? -------C-Tell you how long to take the medica~ion?----------What were his instructions? -----D-Tell you when {what hours ofthe<i'ayj-you.--snoulCI-take the medication? .(What were his instruction?: _=::----- FIGURE PHARMACISTS CHECK LIST 1- '?'bat medication is to be used for _ __ 2- Name of medication --- 3- How to use the medication__ ___ 4- When to use the med.ication_ _ 5- How long to use the medication____ 6- Side reactions the patient might expect_____ 7-What to avoid in activities such as driving, working machinery, eating, drinking, and taking of other drugs ____ _ 8-0.T.C. remedies to avoid if any__ ___ FIGURE SA1'vlPLE OF WRITTEN INSTRUCTIGN F0R PATIENT READ THESE INSTRUCTIuNS CAREFULIJY IV While taking the medication checked below, one or more of the following foods should be taken daily, especially if you are not supplied with a potassium supplement: apricots, bananas, cantaloupe, orange and grapefruit juice, peaches, prunes, and raisins: _____ Ethacrynic acid ---- Furosemide ___ --Hydrochlorothiazide _____ Hydrocortisone ____-Prednisolone This illustration is one of a set of in structions developed by James D. McMahon Ph. D. Professor of Pharmacy Administration, University of Utah, which was used in this study. .1.-