Drug Inf~~nnnIron Jounral. V d 33. pp. 921-931. 1999
`Printed in fhe USA. All rights reserved.
`
`0092-8615199
`Copyright 8 1999 Drug Information Association Inc.
`
`IMPACT OF GEOGRAPHIC
`AND CROSS-CULTURAL DIFFERENCES
`ON SPONTANEOUS ADVERSE
`EVENTS REPORTING
`
`HSIAO-HUI Wu, MS
`Eli I.illy Taiwan. Taipei, Taiwan
`MAN FUNG, MD, MBA, FACP, KEN HORNBUCKLE, DVM, MPH, PHD,
`AND EDMUNDO MUNIZ, MD PHD
`Worldwide Pharmacovigilance and Epidemiology, I d y Research Lab, Indianapolis. Indiana
`
`A study was conducted to explore the potenticil diflerence in geographic' mid cross-culturul
`variution in safee udverse drug reuction reporting. Attempts n'ere made in the study
`design to minimize genetic dflerences, market representution. age. and gender CIS con-
`founding fiictors in affecting the findings. Six thousand spontaneous reports from five
`countries ( Frunce, Germany, United Kingdom, United Stutes, and Canudu) were randomly
`generated from our spontaneous safety dutabase which covered u 15-year reporting
`period (March 1. I983 to December 31, 1997). The reports were then reviewed and
`compared ugainst each cvuntrv for severul factors of interest. Despite some limitations,
`there are interesting observations from the current study that are worth further review.
`For example, German reporis seem more likely to be serious. related to death und life-
`threatenin.? events. and muy nrirrrmt more attention. The Cunudirm and American reports
`hod the highest numbers in the catego? of luck of drug eflect. When .further analyzed
`using continent ruther than coutitty u s a base. Europe was consistently higher in terms
`of number qf serious reports. deatidlife-threatening events. and cases qf overdose. The
`current study is limited to our experience and ,further research by other investigators to
`confirm these findings is wurranted.
`
`Key Words: Spontaneous reporting system; Pharmacovigilance; Cross-cultural differ-
`ences: Safety adverse event reporting
`
`INTRODUCTION
`THE SPoNTANEoUS ADVERSE event re-
`porting system is a widely used and cost-
`effective method to detect adverse drug reac-
`tions in the postmarketing phase of a drug's
`
`Reprint address: Dr. Man Fung. Worldwide P h m a c o -
`vigilance & Epidemiology, I,illy Research Laboratories.
`DC 2531, Indianapolis. IN 46285. E-mail: fung_rn@
`liIly.com.
`
`development (1-9). Since its inception about
`three decades ago, the system has gained
`wide p o p u ~ a ~ t y and is now a standard in
`the pharmaceutical industry for detecting and
`monitoring potential drug safety issues (10-
`13). Many of the regulatory agencies in the
`world have statutory requirements for such
`a system and at times require postmarketing
`surveillance as a criterion for approval of
`certain products in the market (196).
`Although the system has great potential.
`
`92 I
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 1
`
`

`

`922
`
`Hsicro-hui Wu, Mcrn FunK. Ken Hornbuckle. and Edmundo Muniz
`
`there are some factors that may limit its use-
`fulness. For example, the type of the reports
`received as well as the quality of these re-
`ports are important (2.1417). In addition to
`small variations in regional regulatory re-
`quirements in data collection, there are subtle
`differences due to cultural perceptions of se-
`riousness and different considerations of
`whether some reports are worth reporting.
`In order to explore whether cross-cultural
`and geographic differences have any impact
`on safety event reporting, we conducted a
`study of our safety database. The major ob-
`jective of our study is to focus on geographic
`and cross-cultural differences in safety event
`reporting other than genetic differences. As
`far as we know, although there were small
`scattered reports in the literature that exam-
`ined cultural or geographic influences on
`safety event reporting, this was the first
`large-scale systematic study conducted to ex-
`plore this issue.
`
`METHODS
`Twelve hundred spontaneous health care pro-
`fessional reports each from five industrial-
`ized countries in Europe and North America
`were randomly generated from our spontane-
`ous safety database, which covered a 15-year
`reporting period (March 1, 1983 to Decem-
`ber 31, 1997) with 135 different postmar-
`keted drugs in 42 therapeutic classes. The
`countries chosen were the United States,
`Canada, the United Kingdom, France, and
`Germany.
`These five countries were selected based
`on two reasons:
`
`I. Their populations were primarily of Cau-
`casian genetic origin. Thus, unlike com-
`paring Asians with Europeans, differences
`due to genetic composition variation can
`be minimized, and
`2. These five countries were all major phar-
`maceutical markets in which our company
`has an extensive presence. Thus, differ-
`ences due to variation in product availabil-
`ity and market penetration can also be min-
`imized.
`
`In order to minimize potential bias due to
`differences in age among the cohorts se-
`lected. the I200 reports of each country were
`stratified using age as a factor with four dif-
`ferent age groups identified. The groups were
`pediatric/adolescents (< 18 years old; 300 re-
`ports), young adults ( 18-40 years old; 400
`reports), middle age (41-60 years old; 300
`reports), and elderly (&O years old; 200 re-
`ports). The number of reports generated for
`each age group was slightly different in an
`attempt to mimic the normal age distribution
`of the general population. In addition, the
`effect of gender influences on the reports
`was eliminated by stratifying a roughly equal
`number of male to female reports in each
`age group.
`The 6000 randomly generated reports
`from these countries were then reviewed and
`compared against each other in terms of the
`most commonly identified issues in safety
`adverse event reporting. They were the num-
`ber of deathdlife-threatening cases, number
`of serious events, overdose. drug-interaction,
`allergy/anaphylaxis, addiction/abuse, lack of
`drug effect; and the most common body sys-
`tems involved. The comparison was first per-
`formed using all 1200 reports as a whole for
`each country. A subgroup analysis was then
`conducted for each age group of each country
`against one another. Lastly, an intercontinen-
`tal comparison between European and North
`American countries as a group was per-
`formed.
`Statistical analysis was performed as fol-
`lows: Pearson’s Chi Square test or two-sided
`Fisher’s Exact test was conducted for all ad-
`verse events by country for each of the four
`age groups, as well as for all adverse events
`for all age groups combined to test the hy-
`pothesis of equal percentages of these se-
`lected events reported in the five industrial-
`ized countries. Fisher’s Exact test was used
`if the sample size was smaller than five
`counts in some of these analyses. The Coch-
`ran-Mantel-Haenszel (CMH) test was per-
`formed to test for association between coun-
`try or continent (Europe and North America),
`adverse events after adjusting for age group
`effect, as well as the most common body
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 2
`
`

`

`Geographic and Cross-Cultural Differences and Adverse Event Reporting
`
`923
`
`involvement analysis. Statistical
`system
`Analyses Software (SAS) was used to per-
`form all statistical analyses.
`
`RESULTS
`Among the 6OOO spontaneous reports ran-
`domly generated for this study, about one-
`quarter (25.9%) were serious. (“Serious” is
`defined as the commonly accepted regulatory
`criteria including death, life-threatening con-
`ditions, or events required hospitalization,
`major interventions, or involved disability.
`carcinogenicity, teratogenicity, overdose, or
`other clinical serious conditions based on ei-
`ther the reporter or the company’s assess-
`ment.) As for an individual comparison of
`the five countries. German reports had the
`highest number of serious cases (35.3% of
`the 1200 reports), followed by France (3 1.9%).
`with the lowest number of serious cases re-
`ported by Canada (20.3%). Similarly, deaths
`and life-threatening cases were most com-
`monly reported by Germany (12.8%). fol-
`lowed by France (8.3%) with the lowest num-
`ber by the United States (4.9%). As for
`overdose cases, Germany also had the high-
`est number of reports (8.3%). followed by
`France (8.0%). Figure 1 denotes the mean of
`each country across all age groups in terms
`
`of percentage of serious reports, deathflife-
`threatening cases. and overdose cases. Table
`1 displays the breakdown of all selected ad-
`verse events by country and age groups as
`well as the p-values of the statistical analysis.
`Of note is the fact that most of the p-values
`reach statistical significance.
`When serious reports were analyzed using
`the four stratified age groups as a factor, how-
`ever, the observation was similar but not
`identical. Serious reports were highest for
`France for the e l 8 years age group (35.7%
`of the 300 reports), but Germany remained
`the highest for the other three age groups (ie.
`> 18 years old). Germany was also the highest
`for the deathnife-threatening cases for all age
`groups.
`Overdose reports were also quite common
`in this study with about 6.3% of all 6OOO
`reports involving overdose. As discussed be-
`fore, the greatest number of overdose cases
`was again reported by Germany with 8.3%
`of its 1200 randomly generated samples be-
`ing overdose cases. The lowest was Canada
`where only about half as many reports were
`overdoses (4.0%). When broken down by age
`groups, however, no consistent trend as pre-
`viously observed was seen. The highest over-
`dose cases reported were by the United King-
`dom in the pediatridadolescents group (9.7%).
`
`FIGURE 1. Percent of serious events by country for all age^ groups.
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 3
`
`

`

`0.884
`
`0.4%
`0.0%
`0.7%
`0.3%
`0.7%
`0.3%
`
`0.838
`
`0.4%
`0.3%
`0.8%
`0.3%
`0.3%
`0.3%
`
`0.040
`
`0.2%
`0.0%
`1 .O%
`0.0%
`0.0%
`0.0%
`
`0.1 60
`
`1.9%
`3.3%
`1 .O%
`1 .7%
`1 .O%
`2.7%
`
`0.349
`
`2.0%
`2.0%
`3.3%
`2.0%
`1.5%
`1.3%
`
`0.035
`
`1 .3%
`1 .O%
`2.7%
`1 .7%
`0.0%
`1 .O%
`
`Addiction/
`
`Abuse
`
`Interaction
`
`Drug
`
`0.003
`
`1.7%
`1 .7%
`0.3%
`2.7%
`0.3%
`3.7%
`
`0.279
`
`1 .3%
`1.3%
`1 .O%
`2.5%
`0.8%
`1 .O%
`
`0.059
`
`3.2%
`3.0%
`3.0%
`5.7%
`1.3%
`3.0%
`
`0.037
`
`2.5%
`3.0%
`1 .3%
`2.7%
`1 .O%
`4.7%
`
`0.01 8
`
`3.0%
`4.3%
`1 .5%
`3.0%
`1.5%
`4.5%
`
`0.284
`
`3.3%
`4.0%
`3.0%
`3.0%
`4.7%
`1.7%
`
`0.001
`
`5.8%
`4.3%
`3.3%
`10.3%
`3.7%
`7.3%
`<0.001
`9.5%
`7.8%
`9.5%
`8.3%
`16.3%
`5.5%
`<0.001
`5.4%
`3.7%
`9.7%
`7.0%
`6.0%
`0.7%
`
`<0.001
`9.1 Yo
`5.0%
`2.7%
`17.3%
`10.7%
`10.0%
`
`0.144
`6.1 Yo
`4.5%
`5.3%
`8.5%
`5.3%
`6.8%
`<0.001
`4.4%
`2.3%
`2.3%
`8.0%
`6.7%
`2.7%
`
`threatening Overdose Drug Effect Anaphylaxis
`Death/Life-
`
`Allergy/
`
`Lack of
`
`<0.001
`
`23.9%
`22.7%
`11 .7%
`37.0%
`32.0%
`16.0%
`
`<0.001
`
`21 .8%
`18.3%
`18.5%
`30.0%
`24.3%
`18.0%
`
`<0.001
`
`27.1%
`18.7%
`28.7%
`3 1 .3%
`35.7%
`21.3%
`
`p-valuea
`MEAN
`United States
`United Kingdom
`Germany
`France
`Canada
`p-valuea
`MEAN
`United States
`United Kingdom
`Germany
`France
`Canada
`p-valuea
`MEAN
`United States
`United Kingdom
`Germany
`France
`Canada
`
`Serious
`
`Country
`
`41-60 (N = 300)
`
`18-40 (N = 400)
`
`<18 (N = 300)
`
`Percent of Selected Adverse Events by Country For Different Age Groups
`
`TABLE 1
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 4
`
`

`

`3 a
`
`4
`D
`0
`-.
`,
`(x, a
`9
`
`'"
`
`* 6
`5 -
`3
`-
`2
`z
`- -
`f:
`2
`
`i:
`
`bCochran-ManteCHaenszel test was used (df = 4) in testing the association between country and event afler adjusting for age group effect.
`"Pearson's Chi Square test (df = 4) or two-sided Fisher's Exact test was used.
`
`0.047
`
`0.341
`
`<0.001
`
`<0.001
`
`<0.001
`
`<0.001
`
`<0.001
`
`0.104
`
`0.3%
`0.1%
`0.7%
`0.2%
`0.3%
`0.2%
`
`n/a
`
`0.0%
`0.0%
`0.0%
`0.0%
`0.0%
`0.0%
`
`0.341
`
`1.9%
`1 .a%
`2.5%
`2.0%
`1.5%
`1.5%
`
`0.050
`
`2.4%
`0.5%
`3.0%
`3.0%
`4.5%
`1 .O%
`
`<0.001
`
`1.9%
`1 .two
`1.2%
`3.3%
`0.7%
`2.5%
`
`0.009
`
`1 .4%
`1 .5%
`0.0%
`2.5%
`0.0%
`3.0%
`
`<0.001
`
`3.1%
`4.3%
`2.0%
`2.7%
`2.0%
`4.3%
`
`<0.001
`
`3.7%
`6.5%
`2.5%
`1.5%
`0.5%
`7.5%
`
`<0.001
`6.3%
`4.7%
`6.6%
`8.3%
`8.0%
`4.0%
`<0.001
`2.1%
`0.5%
`1 .O%
`7.0%
`1 .O%
`1 .O%
`
`<0.001
`7.7%
`4.9%
`5.3%
`1 2.8%
`am0
`7.2%
`0.004
`
`13.7%
`9.5%
`13.5%
`22.0%
`13.0%
`10.5%
`
`<0.001
`
`25.9%
`21 .3%
`20.5%
`35.3%
`31.9%
`20.3%
`
`<0.001
`
`35.1 '/o
`29.5%
`25.5%
`49.5%
`41.5%
`29.5%
`
`age group effect)b
`Overall P-value (adjusted for
`p-valuea
`MEAN
`United States
`United Kingdom
`Germany
`France
`All Ages (N = 1200) Canada
`p-valuea
`MEAN
`United States
`United Kingdom
`Germany
`France
`Canada
`
`>60 (N = 200)
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 5
`
`

`

`by France in the young adult age group
`(16.3%). and by Germany for the remaining
`two age groups (10.3% for the 41-60 year
`olds and 7.0% for the >60 year olds. respec-
`tively).
`Figure 2 denotes the mean of each country
`across all age groups in terms of percentage
`of drug interaction, lack of drug effect. drug
`addictiodabuse. and allergy/anaphylaxis re-
`ports. Lack of drug effects was most com-
`monly reported by Canada and the United
`States. Overall, about 3. I % of the 6O00 re-
`ports reviewed involved lack of drug effect
`as an event term and the range was from
`2.0% to 4.3%. Drug interaction. allergy/ana-
`phylaxis, and drug abuse/addiction were the
`three other issues explored in this study. All
`of them were quite rare with an overall 1.9%
`reporting rate for allergy/anaphylaxis. 1.9%
`for drug interaction, and only 0.3% for addic-
`tioddrug abuse among the 6OOO reports re-
`viewed.
`In terms of overall reporting pattern by
`country across all ages, allergy/anaphylaxis
`cases were reported most commonly by Ger-
`many (3.3%) and least often by France
`(0.7%). When comparing the same issue us-
`ing age group as a factor. it was not surprising
`that most allergy/anaphylaxis reports were in
`the pediatric/adolescent age group and the
`highest was the S.7% by Germany. Germany
`
`was also highest in the young adult age group
`for this issue (2.5%). On the other hand. Can-
`ada reported the highest allergy/anaphylaxis
`in middle age (3.7%) and elderly patients
`(3.0%). Table I displays the breakdown of
`these selected adverse events by country and
`age groups as well as the p-values of the
`respective statistical analysis.
`As mentioned earlier. drug interaction was
`not very common with only about I .9% of
`the 6000 reports involving some drug inter-
`action. The range was quite small among
`these five countries with the lowest at 1.5%
`in France and Canada and the highest of 2.5%
`in the United Kingdom. Not surprisingly.
`drug interaction was most commonly seen in
`the elderly age group with the French sen-
`iors having the highest reporting of 4.5%.
`compared t o 0.5% in the United States. The
`United Kingdom was the highest in the other
`two age groups (<I8 and I840 year olds)
`with a reporting rate of 7.7% t o 3.3% among
`its patients. The United States had the highest
`reporting of drug interaction among the mid-
`dle age population (3.3%).
`Addiction/drug abuse was either not a
`major issue or it was being underreported as
`the reporting frequency was very low among
`all five countries. The overall average report-
`ing rate was only 0.3% for all countries. The
`United Kingdom was the highest in the pedi-
`
`CA - Canada; FW - France. OR - Ormany. UK - Unitrd ICmKdom. US - United Stntrs
`
`m
`
`c4
`
`m
`
`UI
`
`FIGURE 2. Percent of other events by country for all age groups.
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 6
`
`

`

`atrichdolescent (1.0%) and young adult
`(0.8%) patients. For the middle age patients,
`the United Kingdom and France were tied at
`0.79. There was no report of addiction or
`drug abuse for the elderly age group for any
`of the five countries. Overall, the results were
`consistent with common knowledge that
`drug abuse is more common among adoles-
`cent and young adult patients.
`Although there were obviously cross-
`country differences in each continent, inter-
`estingly, if the data were contrasted by com-
`paring the mean results of the three European
`and two North American countries, the inter-
`continental difference was also quite obvi-
`ous. The European countries were consis-
`tently higher in terms of the number of seri-
`ous reports. deathAife-threatening events,
`cases of overdose. drug interaction. and ad-
`diction/drug abuse. The only exception was
`lack of drug effect and allergy/anaphylaxis
`cases, which were reported the most by the
`North American countries. Figure 3 shows
`these findings.
`As for the body system involvement of
`the events reported. there were also some
`interesting findings. Across all countries and
`all age groups. nervous (CNSPNS) and der-
`matologic systems were the most common
`body systems involved. The United Kingdom
`reported the highest number of cases of ner-
`vous system events (26.5%) while the United
`States reported the highest number of derma-
`
`tologic events ( 18. l %). Figure 4 displays the
`results of this analysis (For ease of visualiza-
`tion, only the seven body systems with the
`highest percentage of events are shown. The
`number of events for the other body system
`is quite low and is not depicted in the current
`figure for simplicity.).
`When a subgroup analysis was conducted
`for the four different age groups, a slightly
`different picture appeared. While nervous
`system complaints remain the most com-
`monly reported type of events for the young
`adult, middle age. and elderly age groups,
`there was an exception. Dermatologic events
`were the most commonly reported events for
`the pediatric/adolescent age group in four of
`the five countries, except the United King-
`dom where nervous system events remained
`the most common.
`Furthermore. the second most common-
`ly involved body system was also quite dif-
`ferent for each age group. For the pediatric/
`adolescent group, musculoskeletal complaints
`were the second most common events re-
`ported in Canada, United States, and France.
`For the young adult group. metabolichu-
`tritional complaints were the second most
`common category for Germany while uro-
`genital events were the second most com-
`mon category for France. For middle age
`patients, metabolichutritional events were
`the second most common reported com-
`plaints for France.
`
`FIGURE 3. Percent of selected events by continent.
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 7
`
`

`

`928
`826
`
`Hsiuo-Imi Wu. Man F1mg. Ken Hombm‘kle. and Edmundo Mum":
`
`
`
`U
`
`
`
`wu
`
`w5,
`
`no OL
`
`WS
`
`wo
`
`
`
`I
`
`
`
`FIGURE 4. Percent of adverse events by body system for all age groups.
`
`There was no consistent pattern for the
`elderly group. While gastrointestinal com-
`plaints were the second most common cate—
`gory for the Canadian and United Kingdom
`seniors. metabolic/nutritional complaints were
`the second highest for French and United
`States elderly and cardiovascular events were
`the most common among the German geriat-
`ric patients.
`
`DISCUSSION
`
`The main objective of the current study is
`no more than an academic exercise to explore
`the potential subtle geographic and inter-
`cultural differences in safety adverse event
`reporting among different countries. Obvi-
`ously. despite our effort to minimize con-
`founding factors by selecting countries with
`relatively similar genetic make-up. market
`representation of our products. age stratifica-
`tion. and equal gender representation. we ac-
`knowledge the limitation in our inability to
`control other influential factors that may af-
`fect the findings. For example. besides dif-
`ference in attitudes in safety adverse event
`reporting and perception of the relative im—
`portance of an event. there are the obvious
`differences in common lifestyle factors such
`as smoking. alcohol use. exercise. and cer-
`tainly different background incidences in
`
`many diseases such as cardiovascular dis-
`eases. diabetes. cancers. prenatal care. and so
`forth among these countries ( 18). In addition.
`differences in the product portfolios among
`different pharmaceutical companies may also
`affect the outcomes of the study. (Theoreti-
`cally. a slight variation in the policies and
`procedures among individual pharmaceutical
`companies with respect to handling adverse
`event reporting might also have some impact
`on the outcomes of the results. Since compli-
`ance to the regulations is very strict among
`these major regulators. the variation should
`be fairly minor and the impact probably will
`be small.)
`
`Despite these limitations. we decided to
`proceed with the study to serve as an example
`for others to explore this issue. With the ef-
`fort of international harmonization. we are
`
`closer to a more unified safety adverse event
`reporting system in the future (26). Over the
`past few decades. we realize that although
`the regulatory bodies for the five countries
`selected all used a similar spontaneous ad-
`verse event reporting system. there are small
`but definite differences among the United
`States Food and Drug Administration. Cana-
`dian Health Protection Branch. United King-
`dom Medicines Control Agency. French
`Medicine Agency. and German Bundesinsti-
`tut fur Arzneimittel und Medizinprodukte in
`
`Pharmacosmos AIS v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 8
`
`PGR2020-00009
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 8
`
`

`

`Geographic and Cross-Cultural Differences and Adverse Event Reporting
`
`929
`
`some of their ggulations (19-27). In addi-
`tion, the level of socialized medicine prac-
`tices in these countries may also have an
`impact on the frequency as well as the type
`of events reported (28). The extent that hos-
`pitals, clinics, and health care work forces
`are owned or operated by the government
`may potentially affect the reporting rate of
`adverse drug events. Lastly, general patient
`and health care provider attitudes toward ad-
`verse drug events and their general percep-
`tion of the relative importance of certain
`types of events or body system involvement
`will also play a role in affecting the pharma-
`covigilance monitoring system (29,30,3 1).
`The impact of these factors is what the cur-
`rent study attempts to explore.
`Due to time and resource limitations, only
`6000 randomly generated spontaneous health
`care professional reports were reviewed.
`Nevertheless, there were interesting results
`observed. For example, about one-quarter of
`all reports were serious and Germany consis-
`tently contributed to most of these cases, in-
`cluding death and life-threatening cases. On
`the other hand, unlike its neighbors France
`and Germany, the United Kingdom was con-
`siderably lower in this category, suggesting
`a difference in either the natural occurrence
`of these events or a different reporting phi-
`losophy among citizens of these countries.
`Even Germany and France showed some dif-
`ferences regarding serious adverse events
`when the data were further analyzed using
`age stratification. Germany had the highest
`number of serious cases for all adult age
`groups from 18 and older but France reported
`the highest number of serious cases for the
`pediatridteenage age group. Thus, one might
`speculate that although Germans generally
`were very concerned about seriousflife-
`threatening events, the French seemed to
`show more concern in their pediatric popula-
`tion (assuming the French children are not
`more susceptible to adverse drug reactions
`than German children). Obviously, these are
`only speculations, but they may stimulate
`further research into the subject of social,
`cultural, and behavioral differences in safety
`adverse event reporting. In all age groups, the
`
`United Sates was in the middle for reporting
`regarding serious adverse events.
`The overdose category was less impres-
`sive in denoting any consistent trend. The
`highest number of overdose cases reported
`were by the United Kingdom in the pediatric/
`adolescents group, by France in the young
`adult age group, and by Germany for the
`remaining two age groups. Canada was the
`lowest in this category in the overall report-
`ing rate.
`Despite being in the middle for reporting
`in most other categories, the United States
`and Canada have the highest reporting rate
`for lack of drug effect. One potential specula-
`tion is that patients (or sometimes physicians
`and pharmacists) in North America were
`generally quite concerned about efficacy of
`the drugs they take. When the desired effect
`did not occur, they would not be hesitant
`to inform the manufacturers or government
`agencies. Again, this remains a speculation,
`but will certainly be worth further pondering
`to account for such observation.
`Drug interaction, allergy/anaphylaxis, and
`addictiorddrug abuse were quite rare and the
`sample size may not be large enough to dem-
`onstrate a significant difference among these
`countries. Nevertheless, our findings were
`consistent with the common medical knowl-
`edge that allergy/anaphylaxis is more com-
`mon in pediatric/adolescents, drug interac-
`tion is more likely in the elderly (due to
`polypharmaceutical use), and drug abuse/ad-
`diction is more common in the adult age
`group. As for cross-country differences, al-
`lergy/anaphylaxis was reported most com-
`monly by Germany (3.3%) and the least by
`France (0.7%). On the other hand, the French
`elderly had the highest reporting rate for drug
`interaction. Adolescents and young adults in
`the United Kingdom had the highest report-
`ing rate of drug addictiodabuse, although
`the overall reporting rates for all countries
`(including the United Kingdom) regarding
`this issue were all quite low.
`When the data were analyzed using an
`inter-continent comparison as an approach,
`there were also some interesting findings.
`Despite obvious cross-country variation, the
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 9
`
`

`

`930
`
`Hsiao-hui Wu, M a n Fung, Ken Hornbuckle. and Edinundo Muniz
`
`European countries were consistently higher
`in terms of the number of serious reports.
`death/life-threatening events. cases of over-
`dose, drug interaction, and addiction/drug
`abuse. The only exception was lack of drug
`effect and allergy/anaphylaxis which was re-
`ported more in North America. There are
`many possible explanations for this observa-
`tion: differences in drug utilization or the
`basic health of the people in the continent,
`and so forth. One may hypothesize, however,
`that perhaps people in the European countries
`may be more earnest in terms of their atti-
`tudes toward adverse drug reaction than their
`North American counterparts. Obviously, this
`is only a speculation and other evidence will
`be needed to substantiate such a bold state-
`ment. Unfortunately, because of the sample
`size and infrequent reporting rates, the latter
`three categories (allergy/anaphylaxis, drug
`interaction, and addiction/drug abuse) did not
`reach statistical significance in this inter-con-
`tinental comparison.
`It is also not surprising that dermatologic
`(eg, rash, itching) and nervous system events
`(eg, headache. drowsiness, numbness) were
`the most common reported complaints. There
`were also some unique findings. Although
`urogenital complaints generally were quite
`rare, there was a very high reporting rate
`(1 3.8%) for France in the adult (1 8-44 years
`old) age group. When the second most com-
`mon involved body system was explored
`(other than dematologic and nervous sys-
`tem), musculo-skeletal complaints were sec-
`ond highest in Canada and France for the
`pediatric/adolescent population. The exact
`implication of these findings is unclear.
`
`CONCLUSION
`We have attempted in this study to identify
`whether cross-cultural or geographic differ-
`ences in safety event reporting exist other
`than genetic differences. Five countries with
`similar genetic make-up as well as market
`representation were selected and stratifi-
`cation was employed to minimize age and
`gender differences as a confounding factor.
`Despite some limitations. we were able to
`
`demonstrate some interesting findings across
`different countries and
`in different age
`groups and offer some potential explanations
`for the observations. Overall, the current
`study is limited to our experience and further
`investigation by other researchers to confirm
`these findings is warranted.
`
`Acknorr./edK,,ie,its-The authors would like to thank Mr.
`Jeffrey Wang for the statistical analysis of the data and
`Ms. Kathryn Li for her helpful comments on the manu-
`script.
`
`REFERENCES
`
`I . Johnson JM. Tanner LA. Postmarketing surveil-
`lance: curriculum for the clinical pharmacologist.
`Part I[: Clinical and regulatory considerations. JClbi
`Phunnacol. 199333: 10 15- 1022.
`Piazza-Hepp 'ID. Kennedy DI.. Reporting of ad-
`2.
`verse events to MedWatch. Am J Heulrh-System
`Phunw 199552: 14361339.
`3. Rossi AC. Knapp DE. Discovery of new adverse
`drug reactions: a review of the Food and Drug Ad-
`ministration's spontaneous reporting system. JAMA.
`1984:252: 1030-1033.
`4. Sills JM. Tanner JA. Milstien JM. Food and Drug
`AdlniniStrAliOn monitoring of adverse drug reac-
`tions. Atti J Hosp Phorm. 1986;43:2764-2770.
`5. Strom BL. Melmon KI.. Can postmarketing surveil-
`lance help to effect optimal drug therapy'? JAMA.
`1979;242:242&2423.
`6. Lortie FM. Postmarketing surveillance of adverse
`drug reactions: problems and solutions. Canudiun
`Med A.ssoc J. 1986: 135:27-32.
`7. Moore N. Riour M. Paux G . et al. Adverse drug
`reaction monitoring: doing it the French way. Lancet.
`l985:2: 1056-1058.
`8. LitovitL T.'Ihe 1ESS database. Use in product safety
`assessment. Drug Sufcry. 1998: I8:9-19.
`9. Finney DJ. The detection of adverse reactions to
`therapeutic drugs. Srut Med. 1982:l: 153-161.
`10. Rossi AC. Bosco 1.. Faich GA. Tanner LA. The
`importance of adverse reaction reporting by physi-
`cians. JAMA. 1988:259: 1203-1204.
`I I . Clark JA. Zimmerman HJ. Tanner LA. Labetalol
`hepntotoxicity. Ann Intern Med. 1990:113:21@213.
`12. Green L. Clark J. Muoroquinolones and theophylline
`toxicity: Norfloxacin. JAMA. 1989:262:2383.
`13. J o h n H. Tanner LA, Green L. Adverse reaction
`reporting of interactions between warfarin and fluor-
`oquinolones. Arch Infern Med. 1991;151:1003-
`IOW.
`14. Belton KJ. Attitude survey of adverse drug-reaction
`reporting by health care professionals across the Eu-
`ropean Union. The European Pharmacovigilance Re-
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1089 - Page 10
`
`

`

`Geographic and Cross-Cultural Differences and Adverse Event Reporting
`
`931
`
`search Group. European J Clin Phurmacol. 1997;
`52:423427.
`15. Cosentino M, Leoni 0, Banfi F, Lecchini S, Frigo
`G. Attitudes to adverse drug reaction reporting by
`medical practitioners in a Northern Italian district.
`Pharmacolog Res. 1997;35:85-88.
`16. Bateman DN, Sanders GL, Rawlins MD. Attitudes
`to adverse drug reaction reporting in the Northern
`Region. Br J CIin Pharmacol. 1992;34:421426.
`17. Generali JA, Danish MA, Rosenbaum SE. Knowl-
`edge of and attitudes about adverse drug reaction
`reporting among Rhode Island pharmacists. Ann
`Pharmacother. 1995;29: 365-369.
`18. Brenner MH. Economic change, alcohol consump-
`tion and heart disease mortality in nine industrialized
`countries. Soc Sci Med. 1987;25:119-132.
`19. Griffin JP. Survey of the spontaneous adverse drug
`reaction reporting schemes in fifteen countries. Br
`J Clin Pharmac. 1986;22(supplement):83S-I00S.
`20. Melnychuk D, Moride Y, Ahenhaim L. Monitoring
`of drug utilization in public health surveillance activ-
`ities: a conceptual framework. Can J Public Health.
`(Revue Canadienne de Sante Publique) 1993;84:
`45-49.
`21. Albengres E. Features of the French postmarketing
`drug surveillance system. Application to cutaneous
`effects of nonsteroidal antiinflammatory drugs. J
`Rheumatol. 1988;17(supplement):20-23.
`22. Albengres E, Gauthier F, Tillement JP. Current
`French system of post-marketing drug surveillance.
`Int J Clin Pharmacol, Therapy, Toxicol. 1990;28:
`3 12-3 14.
`
`23 Kapp JF, Zentgraf R, Widmer A, Schopf E. A need
`to intensify drug surveillance in Germany. Klinische
`Wochenschrif. 1991;69:775-779.
`24. Bem JL, Breckeridge AM, Mann Rd, Rawlins MD.
`Review of yellow cards (1986): report to the commit-
`tee on the safety of medicines. Br J Clin Pharmac.
`1988;26:679489.
`25. Smith CC, Bennett PM, Pearce Hm, et al. Adver