C l i n i c al
`
`C a r e / E d u c a t i o n / N u t r i t i on
`N AL
`A R T I C LE
`
`Impact of Dosage Frequency on Patient
`Compliance
`
`ARSENIO H.P PAES, PHD
`ALBERT BARKER, PHD
`CARMEN J. SOE-AGNIE, PHARMD
`
`OBJECTIVE — To evaluate the impact of dosage frequency on the compliance of patients
`who receive their medicines from community pharmacies.
`
`RESEARCH DESIGN A ND METHODS — Each month, patients received a supply of
`their medication in a Medication Event Monitoring Systems container, which registered each
`opening of the package. At the end of the study, the patients received a short questionnaire. The
`subjects were 91 diabetic patients using oral antidiabetic agents. Patients taking insulin and
`those who were unable to collect their medicines from the pharmacy were excluded from the
`study. Compliance was denned as the percentage of doses taken during the observation
`period. Another parameter used was compliance with the prescribed regimen, defined as the
`percentage of days in which the number of tablets were taken as prescribed. As a last parame-
`ter, compliance with the prescribed dose intervals was used.
`
`RESULTS — Compliance is influenced by the frequency of doses. The compliance for this
`group of patients is 74.8%, with an average of 79% in the case of a dose once daily and 38% in
`the case of a dose three times daily. The predominant type of noncompliance in all groups was
`dose omissions. However, more than one-third of the patients used more doses than prescribed.
`Overconsumption is a frequently made mistake by patients on a one-dose daily schedule.
`
`and twice daily (28,29,42).
`In older studies, compliance is mostly
`measured by pill counts. This method,
`however, provides incomplete and unreli-
`able information that leads to doubtful con-
`clusions. Advances in computer technology
`have made it possible to adapt miniature
`recording devices to medication dispensers
`to record drug use. This aid provides a
`means to obtain details of patients' behav-
`ior during the day and over long periods.
`What is more important, it also provides
`the possibility to observe partial compli-
`ance and the timing intervals between
`doses. In the past few years, this method
`has been used frequently to study compli-
`ance (10,28,29,34,35,39-50).
`In this study, the compliance of a group
`of NIDDM patients was examined. One of
`the objectives of the study was to investi-
`gate the frequency of noncompliance
`among patients using drugs chronically
`and the relationship between dose fre-
`quency and compliance.
`
`CONCLUSIONS — The reduction of dose frequency may decrease total noncompliance,
`but at the same time, it increases the risk of overconsumption. Reducing the frequency does
`not automatically result in a better therapeutic schedule. The choice of once or twice daily
`should depend on the therapeutic range of the drug.
`
`RESEARCH DESIGN AND
`METHODS
`
`N oncompliance with medication can
`
`be considered one of the most serious
`problems facing health care (1-7).
`The effectiveness of treatment of a disease
`depends mainly on two factors: the efficacy
`of the treatment prescribed and the rate of
`compliance of the patient with this treat-
`ment (8). In many cases, the desired effect
`of drugs is not achieved because they are
`not adequately used (9,10). Problems with
`medication compliance occur more fre-
`quently when patients are older (11,12),
`receive more medication (13-15), have to
`take their medicines regularly (16), and
`over a long period of time (17,18).
`Since poor compliance with drugs pre-
`scribed is prevalent, uncomplicated meas-
`
`ures are needed to improve compliance.
`From a practical point of view, it can be
`expected that a simple treatment regimen
`may improve compliance. Various studies
`have shown a relationship between the
`number of doses to be taken and compli-
`ance (10,19-30), but others provide no
`evidence for such a relationship (31-33).
`The results of these studies are not fully
`consistent, but they provide in general a
`view of a higher compliance with once- or
`twice-daily doses than with three- or once-
`daily doses (10,25,34-38). The results of
`some of these studies do not give any evi-
`dence of a difference between dosages once
`or twice daily (10,25,33,39-41); others
`indicate a clear difference between once
`
`From the Department of Pharmacoepidemiology and Pharmacotherapy, Universiteit Utrecht, Utrecht, The
`Netherlands.
`Address correspondence and reprint request to A.H.P Paes, Department of Pharmacoepidemiology and
`Pharmacotherapy, Universiteit Utrecht, PO. Box 80082, 3508 TB Utrecht, The Netherlands.
`Received for publication 5 December 1996 and accepted in revised form 30 April 1997.
`Abbreviations: MEMS, Medication Event Monitoring Systems.
`
`Population
`Patients were recruited from two commu-
`nity pharmacies, based on their chronic
`use of one of the following antidiabetic
`agents: acarbose, glibenclamide, glipizide,
`metformin, or tolbutamide. Patients taking
`insulin were excluded from the study, as
`were patients who were unable to come to
`the pharmacy. The only changes made to
`the prescribed regimens were initiated by
`the prescribing physicians for reasons unre-
`lated to this study. Patients using a weekly-
`dose organizer or another type of special
`container were also excluded.
`As expected, the patients included in
`the trial (n = 91) were relatively old, and the
`group included more women (59.6%) than
`men (40.4%). This is in accordance with
`Dutch morbidity data (51,52). The mean
`age of the female patients was 70.1 ± 10.7
`years, and the mean age of the male patients
`was 67.8 ± 11.2 years. The eldest patient
`was 90 years of age and the youngest was
`45 years of age; both were female. The aver-
`age number of treatment years was 7.1. All
`but two patients lived independently; one-
`
`1512
`
`DIABETES CARE, VOLUME 20, NUMBER 10, OCTOBER 1997
`
`Sawai (IPR2019-00789), Ex. 1051, p. 001
`
`

`

`Paes, Bakker, and Soe-Agnie
`
`third of them lived alone. Almost one-fifth
`of the patients (18.5%) had complaints pos-
`sibly related to complications caused by
`diabetes. One tablet daily was the most
`common dosage schedule (44.0%), twice
`daily was used by 39.6% of the patients,
`and three times daily was the least common
`regimen (16.5%).
`
`Research design
`A total of 91 patients received a 30-day sup-
`ply of their medication each month in a
`Medication Event Monitoring Systems con-
`tainer, which was the only modification in
`their normal drug supply This was done for
`6 months. Patients were asked to keep their
`medication in this container, not to transfer
`their medication to another container, and to
`return it for each refill. At the start of the
`study, patients were asked to take part in a
`research project on a new type of container.
`Two patients refused to enter the study All
`enrolled patients agreed with the procedure.
`At the end of the study, all patients received
`a questionnaire about their actions in the
`case of forgetting to take a tablet and about
`the way they normally prepare and take their
`medication, what they do if a dose is missed,
`how long they receive treatment, etc. Also,
`the working of the containers was explained.
`
`Measurement of compliance
`During the study, three different methods
`were used to measure the patients' compli-
`ance:
`Electronic monitoring with MEMS. MEMS
`(AARDEX, Zurich, Switzerland) is a medica-
`tion bottlecap with a spring-loaded device,
`which, when opened, triggers a switch con-
`nected to a microprocessor that records the
`date and time of opening. Its electronic
`memory also stores information about the
`drug (name and dosage) and the patient
`(patient number). Sequential openings—
`oocasions when containers are opened more
`than once in a brief period—are reported as
`false ones. Data were collected from the
`monitors by connection to a microcomputer.
`MEMS generates an objective simple report
`of the patients dosing record continuously
`since the last dispensation.
`Pill count. Each time the patient came
`back to the pharmacy for his or her pre-
`scription refill, the number of pills left in
`the container was counted by the techni-
`cian or pharmacist who prepared the refill.
`Pharmacy records. Date of refill, dosage,
`number of tablets, and the theoretical date
`when next refill would take place were
`registered at
`the pharmacy. (In The
`
`Table 1—Compliance and dosage
`
`Dosage
`
`Compliance (%)
`
`Range(%)
`
`98.7 ±18.6
`Once daily
`83.1 ±24.9
`Twice daily
`65.8 ±30.1
`Three times daily
`Data are means ± SD, unless otherwise indicated.
`
`19-123
`9-109
`7-102
`
`95% CI
`
`92.8-104.7
`74.7-91.5
`49.1-82.5
`
`Netherlands, patients usually go to the same
`pharmacy for each prescription they get.)
`Data from the questionnaires were
`used to interpret the data from MEMS. The
`questionnaire contained only a few ques-
`tions about baseline data, such as the num-
`ber of years the medicines had been used.
`It also contained data about the patients
`habits in dealing with his or her medicine
`(e.g., storing in other containers) and to
`control if they knew their dose schedule.
`The mean number of days registered per
`patient was 154.8 ± 54.4.
`
`Compliance
`Compliance has been defined as "the extent
`to which the time history of drug adminis-
`tration corresponds to the prescribed regi-
`men" (53). It includes both dose-taking and
`dose-timing aspects. In this study, compli-
`ance was defined as the number of doses
`taken (number of container openings)
`divided by the prescribed number of doses
`during the observation period multiplied by
`100. This parameter includes partial com-
`pliance as well as overconsumption.
`Another parameter used was the regi-
`men compliance. This was defined as the
`percentage of days in which the prescribed
`dose regimen (twice, three times, or once
`daily) was taken as prescribed.
`Deviation from the prescribed dosing
`intervals was another aspect of compliance.
`This was defined as the percentage of pre-
`scribed doses taken within 25% of the pre-
`scribed
`interval. For example, if the
`prescribed regimen was twice daily, a time
`period of 9-15 h between the two doses
`met this criterion. For once-daily prescrip-
`tions, such a time interval was 18-30 h and
`for three times daily it was 6-10 h.
`Refill compliance was calculated with
`data in the medication history of the
`patient at the pharmacy. In this study, only
`too-late refills were recorded. Early refills
`can have other reasons apart from over-
`consumption.
`
`Statistical analysis
`Analysis of variance (and Tukey's honestly
`
`significant difference [HSD] test) and Stu-
`dent's t test with Bonferroni multiple com-
`parisons were used for group comparisons.
`When nonparametric tests were required,
`the Kruskal-Wallis test was used. A value of
`<0.05 was considered statistically signifi-
`cant. Effect size statistics were calculated
`with Cohen's d.
`
`RESULTS— The mean percentage of
`prescribed doses taker was 74.8 ± 26.0%,
`with a range from 7 to 123%. The mean
`percentage of days in which the doses were
`taken as prescribed (regimen compliance)
`was 67.2 ± 30.0%, with a range from 0 to
`97%. These percentages are a little higher
`than the results obtained by other studies
`on compliance with chronic drug use
`(1,7,18,43,54,55), but are lower than those
`of other studies of compliance and oral
`antidiabetic agents (56).
`In this study, the data showed a clear
`relationship between compliance and the
`number of daily doses. This in accordance
`with other studies (10,19-27,30). The
`compliance increases with a reduction of
`the number of doses.
`Table 1 lists the compliance percent-
`ages as measured with MEMS data with
`once, twice, or three times daily dosage reg-
`imens. Table 2 shows the compliance with
`the prescribed regimen. In both cases, the
`differences among the three dosage regi-
`mens are significant (P < 0.05).
`Also, the effect size statistics show a dif-
`ference between
`the dosage regimens
`(Table 3). According to Cohen's interpreta-
`tion of these descriptions (57), the differ-
`ences in compliance between once and
`twice daily and between twice and three
`times daily can be considered medium and
`between once and three times daily as large.
`The compliance with the prescribed regi-
`men also shows a difference: medium
`between once and twice daily and large
`between once and three times daily and
`between twice and three times daily.
`The higher compliance with once-daily
`dosage schedules can be understood from
`observing the chronology of the different
`
`DIABETES CARE, VOLUME 20, NUMBER 10, OCTOBER 1997
`
`1513
`
`Sawai (IPR2019-00789), Ex. 1051, p. 002
`
`

`

`Dosage frequency and patient compliance
`
`Table 2—Compliance with prescribed regimen
`
`Dosage
`
`Compliance (%)
`
`Range(%)
`
`Once daily
`79.1 ± 18.8
`Twice daily
`65.6 ±29.7
`Three times daily
`38.1 ±35.9
`Data are means ± SD, unless otherwise indicated.
`
`8-96
`0-96
`1-97
`
`95% Cl
`
`73.2-85.0
`56.5-74.1
`21.2-55.0
`
`Dose omissions were the predominant
`type of noncompliance in all groups. But
`over one-third (37.4%) of the patients used
`more doses than prescribed. The mean
`number of days patients took more than
`prescribed was 7.7 ± 8.9 (mean number
`registered, 154.8 days). The mean number
`of extra doses used was 11 ± 16.6 (mean
`number registered, 306.5 doses). Overcon-
`sumption increased with a decrease of the
`number of doses. Almost 40% of the
`patients with a once-daily dosage regimen
`were taking more tablets than prescribed.
`Almost always the extra dose was taken
`later in the day. With twice or three times
`daily dosages, the percentage of overcon-
`sumption was much lower (11.1 and
`13.3%, respectively). When only the num-
`ber of tablets taken is used as a measure of
`compliance, it gives a wrong impression of
`the compliance of these patients because it
`disregards the influence of overconsump-
`tion.
`Patients were questioned about their
`habits when they expect to overeat. Only
`one patient indicated that he adjusted his
`medication to his eating behavior. Over-
`consumption, thus, must be regarded as
`inadvertent. Over one-third (34.8%) of the
`noncompliance resulted in a 24-h period
`without any dose having been taken. This
`percentage of a 24-h period without thera-
`peutic covering was higher (P < 0.05)
`among patients with a once-daily regimen
`than with twice (d = 0.71) or three times {d
`= 1.12) daily regimens. Between the last
`two, there is no significant difference in the
`number of 24-h periods without therapeu-
`tic coverage.
`
`patients. The first tablet was mostly taken
`very regularly (probably during breakfast).
`The taking of the second and especially the
`third tablets becomes more irregular.
`The mean opening times of the con-
`tainer for patients on a twice-daily dosage
`(if both tablets are used) were 8:24 A.M.
`(±2.1 h) and 6:24 P.M. (±4.1 h). For
`patients on a three times daily dosage, the
`mean times were 7:48 A.M. (±1.3 h), 1:24
`P.M. (±1.9 h), and 6:06 P.M. (±3.9 h). Figure
`1 shows the most common pattern of tak-
`ing medicines with a dosage of once, twice,
`and three times daily. Also, compliance
`with dose intervals is influenced by the
`number of doses. The differences between
`the three regimens are significant (P <
`0.01) (Table 4).
`The very low compliance with the dose
`interval at a three times daily dosage can
`probably be explained by the relationship
`with meals. Patients are advised to take
`their medicines with food or during meals.
`Using MEMS, we can observe that for an
`important number of people, the time of
`administration is concentrated during a 10-
`to 11-h period (between 8:00 A.M. and 7:00
`P.M.). However, there is a remarkably wide
`interindividual variation within the once-
`daily dosage regimen.
`If we calculate compliance using the
`pill-count data, the percentage of compliant
`patients was 72.5%. The mean number of
`returned tablets was 5.6 ± 16.5. The com-
`pliance of the whole group was 99.85%.
`There are no significant differences among
`the three different dosage regimens. The
`refill compliance (refilling on time) was
`63.6%. If we only consider those patients
`who returned after >6 days noncompliant,
`then the compliance increases to 77.7%.
`The average number of days' delay was
`7.8, with a maximum of 81 days. There are
`no significant differences in refill compli-
`ance among the different dosage regimens.
`Baseline patient characteristics such as the
`number of years the medicines were used,
`age, and sex do not show a correlation
`with compliance.
`
`Differences in "drug holidays" among
`the three groups were less pronounced. A
`drug holiday was defined as ^3 days with-
`out medication. The mean number of drug
`holidays was 2.53 with a regimen of once
`daily, 2.03 with a regimen of twice daily, and
`0.59 with a regimen of three times daily.
`The difference between once and twice
`daily and thrice daily is significant (P <
`0.05; d = 1.71 and d = 1.18, respectively).
`
`CONCLUSIONS— Our investigation
`was a study of compliance with oral antidi-
`abetic agents in three different dosage sched-
`ules. Compliance in our group of patients
`was 74.8%, and compliance with the pre-
`scribed dosage regimen was lower at 67.5%.
`In general, diabetic patients can be con-
`sidered relatively well informed about their
`disease and the need for medication. The
`compliance in this study was higher than
`the results from other studies of chronic
`drug use, but lower than studies of compli-
`ance with oral antidiabetic agents. It seems
`important to mention the dosages in reports
`about compliance because of the influence
`of the number of tablets to be taken daily in
`compliance. There is a significant differ-
`ence in compliance between different
`dosage regimens. There is a difference in the
`number of tablets used and in compliance
`with the prescribed regimen. The pharma-
`ceutical industry suggests that a once-daily
`regimen improves patient compliance. Also,
`some authors recommend that practitioners
`select medications that permit the lowest
`daily dose (58). Earlier, Haynes et al. (59)
`warned against these claims because they
`were not warranted.
`In this study, making use of MEMS, it
`was possible to observe the number and
`moments of (possible) intake. Our results
`show that the compliance, the percentage of
`tablets used, and the number of days in
`which the correct number of doses was
`used increase with the decrease in the num-
`ber of doses to be taken daily. Based on
`these results, it may be recommended to
`prescribe an easy dosage regimen of one
`tablet daily, and some authors do (58).
`
`Table 3—Effect size statistics and dose schedule
`
`Dosage
`
`Compliance
`
`Compliance with prescribed regimen
`
`Once/twice daily
`Once/three times daily
`Twice/three times daily
`Data are Cohen's d statistics.
`
`0.73
`1.35
`0.61
`
`0.61
`1.60
`0.81
`
`1514
`
`DIABETES CARE, VOLUME 20, NUMBER 10, OCTOBER 1997
`
`Sawai (IPR2019-00789), Ex. 1051, p. 003
`
`

`

`Paes, Bakker, and Soe-Agnie
`
`A higher compliance with a once-daily reg-
`imen is accompanied by higher overcon-
`sumption and more 24-h periods without
`therapeutic coverage. For the practitioner,
`this means that he or she has to weigh the
`risks of partial compliance and overcon-
`sumption when choosing a dosage regimen,
`considering, of course,
`that under-
`consumption is more frequent than over-
`consumption. Also, the risk of days
`without therapeutic coverage has to be
`taken into account. The warning of Haynes
`et al. (59) is still valid: A dosage frequency
`of once daily is not always the best choice.
`Generally, studies of patient compliance are
`focused on partial compliance. The results
`of this study support the results of a few
`other studies (33,43,62) that overcon-
`sumption also has to be taken into account.
`Only counting the number of doses con-
`sumed overestimates compliance and does
`not consider the days without therapeutic
`coverage.
`Overconsumption in the group of
`patients with a once-daily regimen can
`probably be explained as the result of the
`insecurity of the patient about his or her
`own consumption behavior. The additional
`intakes occur late on the day when the
`patient probably feels insecure about
`whether he or she has already taken his or
`her medicine or not. In the literature, it has
`been suggested that higher compliance can
`be reached when aids, such as Doset box,
`are used (63-66). They give visual clues
`and show what to discard.
`A higher dosage schedule of three
`times daily is responsible for a higher par-
`tial compliance, but this scheme also shows
`a lower number of daig holidays than the
`other two dosage schedules. Studies of the
`administration of eyedrops show a higher
`compliance with the morning than with the
`evening dose (67,68). In a study of com-
`pliance with anti-epileptics, it was found
`that the first dose was missed more fre-
`quently (69), but in another study, no dif-
`ference was found between the morning
`and evening intakes (50).
`
`Table 4—Compliance with interdose intervals
`
`Dosage
`
`Compliance (%)
`
`Range (%)
`
`Once daily
`Twice daily
`Three times daily
`Data are means ± SD, unless otherwise indicated.
`
`77.7 ±21.1
`40.7 ± 28.2
`5.3 ±5.3
`
`2-98
`0-88
`1-75
`
`95% CI
`
`71.0-84.5
`31.1-50.2
`2.3-8.2
`
`1515
`
`DIABETES CARE, VOLUME 20, NUMBER 10, OCTOBER 1997
`
`o
`
`o
`
`&
`
`o
`
`%
`
`"
`
`O O O
`
`o
`
`o
`
`24.00 -
`
`20.00 -
`
`18.00
`
`-
`
`12.00 -
`
`04.00 -
`
`B
`
`Figure 1—Examples ojonce (A), twice (B), and three times (Q daily dosage regimens.
`
`may be insufficient to overcome the result of
`missed doses (61).
`Our results show that in a group of dia-
`betic patients, overconsumption can also
`occur. Only manipulating the doses would
`not always result in a therapeutic optimum.
`
`However, at the same time, our data show
`that overconsumption is higher with a once-
`daily regimen than with a twice or three
`times daily one. Between the last two, there
`is no significant difference. Some authors
`recommend a twice-daily dosage as more
`reliable in case a dose is missed by the
`patient (60). In support of this practice, the
`results of our study show a significantly
`higher number of 24-h periods without
`therapeutic coverage among patients on a
`once-daily regimen. On the other hand,
`there are no differences in the number of
`drug holidays between once- and twice-
`daily regimens. As has been pointed out in
`the past, the reduction of noncompliance
`obtained by reducing the dosage frequency
`
`Sawai (IPR2019-00789), Ex. 1051, p. 004
`
`

`

`Dosage frequency and patient compliance
`
`As in other studies (45,49,53), our
`study found that the morning dose was
`taken more regularly than the others. A
`possible explanation for compliance with
`the first dose may be the fact that these
`patients are recommended to take their
`medicines with meals, and, of course,
`breakfast is the first meal of the day.
`In summary, the present study sup-
`ports the hypothesis that the prescribed
`dosage frequency influences patient com-
`pliance: the lower the dosage frequency, the
`higher the patients compliance, but also
`the higher the overconsumption. Conse-
`quently, our results do not suggest that a
`once-daily regimen is always the first
`choice. Reducing the frequency of a dosage
`will not automatically result in a better
`therapeutic schedule because of
`the
`increase in overconsumption with a once-
`daily regimen and an increase of the num-
`ber of 24-h periods without medication.
`Another aspect is the interval between
`two doses. Interval compliance is highest
`among patients on a once-daily regimen
`and lowest among those on a three times
`daily regimen. But between once- and
`twice-daily dosages, there are also signifi-
`cant (P < 0.05) differences in compliance
`with interdose intervals (Table 3). This is
`the result of the fact that a second dose is
`taken more irregularly. If interval compli-
`ance is important, then it is important to
`remember that a more complicated dosage
`schedule may result in a higher noncom-
`pliance with interdose intervals.
`
`References
`1. Stunkard AJ: Adherence to medical treat-
`ment: over-view and lessons from behav-
`ioral weight control. ] Psychosom Res
`25:187-197,1981
`2. Levy RA: Failure to refill prescriptions. In
`Patient Compliance in Medical Practice and
`Clinical Trials. Cramer JA, Spilker B, Eds.
`New York, Raven Press, 1991, p. 11-18
`3. Bond WS, Hussar DA: Detection methods
`and strategies for improving medication
`compliance. Am] Hosp Pharm 48:1978-
`1988,1991
`4. Urquhart J: Ascertaining how much com-
`pliance is enough with outpatient antibiotic
`regimens. Postgrad Med J 68 (Suppl.
`3):S49-S59, 1992
`5. Blom ATG, Paes AHP: De therapie-trouw
`kan worden verbeterd [Patient compliance
`can be improved]. Geneesmiddelenbulletin
`26:40-43,1992
`6. Paes A, Cornips J: Gebruik volgens
`voorschrift [Use as indicated]. lntermediair
`16:1-11, 1980
`
`7. Wright EC: Non-compliance: or how many
`aunts has Matilda? Lancet 342:909-913,
`1993
`8. Epstein LH: The direct effects of compli-
`ance on health outcome. Health Psychol
`3:385-393, 1984
`9. Mikael RL, Sharpe T: Patient compliance.
`In Pharmacy Practice. Wertheimer AI, Smith
`MC, Eds. Baltimore, MD, University Park
`Press, 1974, p. 179-194
`10. Cramer JA, Mattson RH, Prevey ML,
`Scheyer RD, Ouellette VL: How often is
`medication taken as prescribed? A novel
`assessment technique. JAMA 261:3273-
`3277,1989
`11. Bockowski JA, Zeichner A: Medication
`compliance and the elderly Clin Gerontol
`4:3-15, 1985
`12. Weingarten MA, Cannon BS: Age as a major
`factor affecting adherence to medication for
`hypertension in a general practice popula-
`tion. Fam Practice 5:294-296, 1988
`13. Evans L, Spelman M: The problem of non-
`compliance with drug therapy. Drugs
`25:63-76, 1983
`14. Owens NJ, Larrat EP, Fretwell MD: Improv-
`ing compliance in the older patient: the role
`of comprehensive functional assessment.
`In Patient Compliance in Medical Practice and
`Clinical Trials. Cramer JA, Spilker B, Eds.
`New York, Raven Press, 1991, p. 107-119
`15. Stuart B, Coulson NE: Dynamic aspects of
`prescription drug use in an elderly popula-
`tion. Health Res 28:237-264,1993
`16. Howie VM, Ploussard JH: Compliance
`dose-response relationships in streptococ-
`cal pharyngitis. Am] Dis Child 123:18-25,
`1972
`17. Luescher TF, Vetter H, Siegenthaler W, Vet-
`ter W: Compliance in hypertension: facts
`and concepts. J Hypertension 3 (Suppl.
`l):3-9, 1985
`18. Hammel RJ: Increased compliance means
`better health for patients, higher profits for
`you. Am Druggist 184:98, 1981
`19. Gatley MS: To be taken as directed. J R Coll
`Gen Pract 16:39-44, 1968
`20. Ayd FJ: Once-a-day neuroleptic and tri-
`cyclic antidepressant therapy. Int Drug Ther
`Newskt 7:33-40, 1972
`21. Ayd FJ: Rational pharmacotherapy: once-a-
`day dosage. Dis Nerv Syst 34:371-378,
`1973
`22. Wandless I, Mucklow JC, Smith A, Prudham
`D: Compliance with prescribed medicines: a
`study of elderly patients in the community J
`R Coll Gen Pract 29:391-396,1979
`23. Fujii J, Akira S: Compliance and compli-
`ance-improving strategies in hypertension:
`the Japanese experience. ] Hypertens 3
`(Suppl.): 19-22, 1985
`24. CockbumJ, Gibberd RW, Reid AL, Sanson-
`Fisher RW: Determinants of non-compli-
`ance with short term antibiotic regimens.
`BrMed] 295:814-818, 1987
`25. Pullar T, Birtwell AJ, Wiles PG, Hay A,
`
`Seely MP: Use of pharmacologic indicator
`to compare compliance with tablets pre-
`scribed to be taken once, twice, or three
`times daily. Clin Pharmacol Ther 44:540-
`545,1988
`26. Sclar DA: Improving medication compli-
`ance: a review of selected issues. Clin Ther
`13:436-440,1991
`27. Nicholas WC, Fisher RG, Stevenson RA,
`Bass JD: Single daily dose of methimazole
`compared to every 8 h propylthiouracil in
`the treatment of hyperthyroidism. South
`MedJ 88:973-976, 1995
`28. Waeber B, Erne P, Saxenhofer H, Heynen G:
`Use of drugs with more than a 24-h duration
`of action.; Hypertens 12:S67-S71,1994
`29. DetryJM, Block P, De Backer G, Degaute JP,
`Six R: Patient compliance and therapeutic
`coverage: amlodipine versus nifedipine
`(slow-release) in the treatment of angina
`pectoris: Belgian Collaborative Group. J Int
`Med Res 22:278-286,1994
`30. Sclar DA, Tartaglione TA, Fine MJ:
`Overview of issues related to medical com-
`pliance with implications for the outpa-
`tient management of infectious diseases.
`Inject Agents Dis 3:266-273, 1994
`31. Widmer RB, Cadoret RJ, Troughton E:
`Compliance characteristics of 291 hyper-
`tensive patients from a rural midwest area.
`JFam Pract 17:619-625, 1983
`32. Leirer VO, Morrow DG, Pariante GM,
`Sheikh JI: Elders' nonadherence: its assess-
`ment, and computer assisted instruction
`for medication recall training. J Am Geriatr
`Soc 36:877-884,1988
`33. KruseW, RampmaierJ, Ullrich G, Weber E:
`Patterns of drug compliance with medica-
`tion to be taken once and twice daily
`assessed by continuous electronic moni-
`toring in primary care. Int] Clin Pharmacol
`Ther 32:452-457, 1994
`34. Kruse W, Eggert-Kruse W, Rampmaier J,
`Runnebaum B, Weber E: Dosage frequency
`and drug-compliance behavior: a compar-
`ative study on compliance with a medica-
`tion to be taken twice or four times daily
`Eur] Clin Pharmacol 41:589-592, 1991
`35. Eisen SA, Miller DK, Woodward RS, Spitz-
`nagel E, Pzybeck TR: The effect of pre-
`scribed daily dose frequency on patient
`medication compliance. Arch Intern Med
`150:1881-1884, 1990
`36. Taggart AJ, Johnson GD, McDevitt DG:
`Does the frequency of daily dosage influ-
`ence compliance with digoxin therapy? Br
`J Clin Pharmacol 1:31-34,1981
`37. Greenberg RN: Overview of patient com-
`pliance with medication dosing: a literature
`review. Ciin Ther 6:592-599, 1984
`38. Farmer KC, Jacobs EW, Phillips CR: Long-
`term patient compliance with prescribed
`regimens of calcium channel blockers. Clin
`Ther 16:316-326, 1994
`39. Kruse W, Koch GP, Nikolaus T, Oster P,
`Schlierf G, Weber E: Measurement of drug
`
`1516
`
`DIABETES CARE, VOLUME 20, NUMBER 10, OCTOBER 1997
`
`Sawai (IPR2019-00789), Ex. 1051, p. 005
`
`

`

`Paes, Bakker, and Soe-Agnie
`
`compliance by continuous electronic mon-
`itoring: a pilot study in elderly patients dis-
`charged from hospital. J Am Geriatr Soc
`40:1151-1155, 1992
`40. Meredith PA: Patient compliance (Letter).
`BMJ 305:1434,1992
`41. Kruse W, RampmaierJ, Ullrich G, Weber E:
`Patterns of drug compliance with medica-
`tions to be taken once and twice daily
`assessed by continuous electronic moni-
`toring in primary care. IntJ Clin Pharmacol
`Ther 32:452-457, 1994
`42. Brun J: Patient compliance with once-daily
`and twice-daily oral formulations of 5-
`isosorbide mononitrate: a comparative
`study. J Int Med Res 22:266-272,1994
`43. Kruse W, Weber E: Dynamics of drug regi-
`men compliance: its assessment by micro-
`processor-based monitoring. Eur J Clin
`Pharmacol 38:561-565, 1990
`44. Rudd P, RameshJ, Bryant-Kosling C, Guer-
`rero D: Gaps in cardiovascular medication
`taking: the tip of the iceberg. J Gen Intern
`Med 8:659-666,1993
`45. Mengden T, Binswanger B, Spuhler T,
`Weisser B, Vetter W: The use of self-meas-
`ured blood pressure determinants in assess-
`ing dynamics of drug compliance in a study
`with amlodipine once a day, morning versus
`evening. ] Hypertens 11:1403-1411, 1993
`46. Averbuch M, Weintraub M, Pollock DJ:
`Compliance assessment in clinical trials:
`the MEMS device. J Clin Res Pharmacoepi-
`demiol 4:199-204,1990
`47. Urquhart J: When outpatient drug treat-
`ment fails: non-complier or non-responder:
`identifying non-compliers as a cost-con-
`tainment tool. Clin Res Reg Affairs 11:19—
`38,1994
`48. Cramer JA: Microelectronic systems for
`monitoring and enhancing patient compli-
`ance with medication regimens. Drugs
`
`59. Haynes RB, Sackett DL, Taylor DW, Roberts
`49:321-327,1995
`RS, Johnson AL: Manipulation of the ther-
`49. Kruse W, Nikolaus T, Rampmaier J, Weber
`apeutic regimen to improve compliance:
`E, Schlierf G: Actual versus prescribed tim-
`conceptions and misconceptions. Clin
`ing of lovastatin doses assessed by elec-
`Pharmacol Ther 22:125-130,1977
`tronic compliance monitoring. Eur J Clin
`Pharmacol 45:211-215,1993
`60. Keen PJ: What is the best dosage schedule
`for patients? JR Soc Med 84:640-641,1991
`50. MallionJM, Meilhac B, Tremel E. Calvez R,
`Bertholom N: Use of a microprocessor-
`61. Levy G: A pharmacokinetic perspective on
`equipped tablet box in monitoring compli-
`medication noncompliance. Clin Pharmacol
`ance with antihypertensive treatment. J
`Ther 54:242-244,1993
`Cardiovasc Pharmacol 19 (Suppl. 2):S41-
`62. Cramer JA, Mattson RH: Monitoring com-
`S48,1992
`pliance with antiepileptic drug therapy. In
`Patient Compliance in Medical Practice and
`51. Herings RMC: Effecten van chronisch/gecom-
`Clinical Trials. Cramer JA, Spilker B, Eds.
`bineerdgebruik vangeneesmiddelen [Effects of
`New York, Raven Press, 1991, p. 123-137
`chronic and combined drug use]. Utrecht,
`The Netherlands, Vakgroep Praktische Far-
`63. Crome P, Akerhurst M, Keet J: Drug compli-
`macie,1989
`ance in elderly hospital inpatients: trial of the
`Dosettbox. Practitioner 30:329-333,1980
`52. Stolk RP, Grobbee DE: Epidemiologie van
`64. Kjellgren KI, Ahlner J, Saljo R: Taking anti-
`diabetes mellitus [Epidemiology of diabetes
`hypertensive medication: controlling or co-
`mellitus]. Diagnose Injormatie en Medische
`operating with patients? Int J Cardiol
`Statistiek (DIMS) 12:4-7, 1992
`47:257-268,1995
`53. Urquhart J: Role of patient compliance i