Chester Clinic

 

Decker.vp:corelventura 7.0

Color profile: Disabled Composite Default screen Neisseria gonorrhoeae isolates in Metropolitan Toronto: Issues of scale, positional accuracy and confidentiality Jody F Decker PhD, Bob Sharpe PhD, Jo-Anne R Dillon PhD JF Decker, B Sharpe, J-AR Dillon. Mapping antibiotic-resistant Neisseria gonorrhoeae isolates in Metropolitan To-
ronto: Issues of scale, positional accuracy and confidentiality. Can J Infect Dis 1997;8(5):273-278. The primary ob-
jective of this paper was to investigate the methodological implications of mapping Neisseria gonorrhoeae using the
partial three-digit postal code instead of the complete six-digit postal code. The reporting locations of N gonorrhoeae iso-
lates submitted from hospitals, doctor's offices, private and provincial laboratories, and sexually transmitted disease
(STD) clinics were used as a model. Specifically, the paper focused on variations in geographical distributions of STD
data when mapped at different aggregations of postal code data and at different map scales. Such variations are of im-
portance to those who analyze the spatial epidemiology of STDs, and the accessibility and use of health care services.
This analysis showed that three-digit postal codes are useful in summarizing overall geographic distributions, but
greatly reduce positional accuracy, which can lead to inaccurate delineation of service areas and populations served. The
six-digit postal code is more appropriate for detailed analysis addressing behavioural issues. The analysis demonstrated
that six-digits can be used without breaching individual confidentiality.
Key Words: Geographic mapping, Neisseira gonorrhoeae
Cartographie des isolats de Neisseria gonorrheæ résistants aux antibiotiques dans la région
de Toronto : problèmes d'échelle, de précision des localisations et de confidentialité
RÉSUMÉ : L'objectif premier de cet article était d'étudier les implications méthodologiques de la cartographie de
Neisseria gonorrheæ à l'aide du code postal partiel à trois caractères plutôt que du code postal complet à six caractères.
La localisation des isolats de N. gonorrheæ signalés par les hôpitaux, les cabinets de médecins, les laboratoires privés et
provinciaux et les cliniques de maladies transmissibles sexuellement (MTS) ont servi de modèles. Plus précisément,
l'article a porté sur les variations de distribution géographique des données sur les MTS lorsqu'elles étaient
cartographiées selon diverses agrégations de données à partir des codes postaux et selon diverses échelles
cartographiques. De telles variations sont importantes pour ceux qui analysent l'épidémiologie géographique des MTS et voir page suivante Department of Geography and Environmental Studies, Wilfrid Laurier University, Waterloo, Ontario; and Laboratory Centre for Disease Control, Ottawa, Ontario Correspondence: Dr Jody Decker, Department of Geography and Environmental Studies, Wilfrid Laurier University, Waterloo, Ontario N2L 3C5. Telephone 519-884-1970 ext 2215, fax 519-725-1342, e-mail [email protected] Received for publication July 30, 1996. Accepted January 14, 1997 Can J Infect Dis Vol 8 No 5 September/October 1997 G: INFDIS 1997 Vol8no5 decker.vp Thu Oct 02 15:18:04 1997 Color profile: Disabled Composite Default screen Decker et al
l'accessibilité des services de santé. Cette analyse a permis de révéler que les codes postaux à trois caractères sont utiles pour résumer les distributions géographiques globales, mais qu'ils donnent une idée considérablement plus vague de la répartition géographique, ce qui peut donner lieu à l'établissement de limites floues quant aux zones de service et aux populations desservies. Le code postal à six caractères est plus précis pour une analyse détaillée portant sur des questions de comportement. L'analyse a démontré que le code postal à six caractères peut être utilisé sans bris de confi-dentialité.
In Canada, the incidence of Neisseria gonorrhoeae has allyTransmittedDiseases(NLSTD),LaboratoryCentreforDis- steadily declined for over a decade; from 1981 to 1993, it ease Control, Ottawa, Ontario. Data included laboratory data decreased from 56,336 to 6820 cases per year (1). However, on antibiotic resistance, as well as specific isolate data (eg, submitting location) which the Ontario Laboratory Services producing isolates of N gonorrhoeae (PPNG), has continued to Branch – Etobicoke had originally given the NLSTD. Only be an ongoing problem (2-4). Isolates of N gonorrhoeae with PPNG, TRNG and PPTRNG isolates of N gonorrhoeae were re- plasma-mediated resistance to tetracycline have increased (5).
corded. These antibiotic-resistant isolates were used because As gonorrhoea persists as a public health problem (6), appro- over 90% of such reported isolates were forwarded to the priate diagnosis and treatment are essential disease control NLSTD as part of a national surveillance program, thus pro- strategies, and the surveillance of antimicrobial susceptibility viding a detailed snapshot of such isolates.
is essential to monitor ongoing and emerging resistance.
Each record in the data set contained a postal code for the Knowledge of the geographic distribution of N gonorrhoeae submitting location from which isolates were sent for analy- can be a useful tool for assessing patterns and their underly- sis. These locations included practitioner's offices; provincial ing processes.
STD clinics; primary medical laboratories, Central Public A cursory examination of the published data shows impor- Health Laboratory, Laboratory Services Branch, Ontario Min- tant geographic variations and a pronounced urban bias in the istry of Health, Etobicoke, Ontario,; or hospitals. Although the distribution of N gonorrhoeae isolates with plasmid-mediated data set initially contained 133 different submitting locations, resistance. The provinces of Ontario and Quebec contained not all had a unique location or postal code. Many of the sub- 92% of total Canadian PPNG cases in 1989, and the large ur- mitting locations were smaller laboratories in larger facilities ban centres in these provinces, Toronto and Montreal, respec- and thus shared postal codes. As a result, the number of sub- tively, accounted for 68.2% and 55.5% of their respective mitting location records with unique six-digit postal codes provincial totals (7). Furthermore, Brown et al (8) reported a was 75 rather than 133. For the purpose of comparison, the high prevalence of PPNG in Ontario, particularly in the Metro- records were further aggregated using only three digits of the politan Toronto area, which also carried plasmid-mediated re- postal codes. This reduced the number of submitting locations sistance to tetracycline (tetracycline-resistant N gonorrhoeae [TRNG]), alone or in combination with PPNG (PPTRNG).
To map their spatial distribution, geographic coordinates Further geographic analysis with more disaggregated data for each submitting location were assigned using the postal may reveal patterns of spatial concentration within the city.
code address. This was accomplished using the postal code The locational data commonly used for this purpose in Canada conversion file of Statistics Canada. For each postal code in are the first three digits of the postal code. Our objective was Canada, this file specifies its precise location, using either lati- to investigate some of the methodological implications of us- tudinal and longitudinal coordinates or the Universal Trans- ing partial three-digit postal code versus the more positionally verse Mercator (UTM) system, established international accurate complete six-digit postal code. In this instance, the systems of specifying point locations on the globe. The postal client's place of residence postal code was not available be- codes consist of six alpha-numeric characters intended to de- cause of confidentiality concerns and is rarely available to re- scribe the destination of each item of mail addressed in Can- searchers analyzing sexually transmitted diseases (STDs), so ada (for example, M2M 2R9). The first character of the postal we mapped submitting locations of isolates in Metropolitan code represents a province or territory or major sector within a Toronto. Using submitting locational data, we demonstrated province (for example, ‘M' represents Metropolitan Toronto).
that methodologically the use of partial three-digit postal The first three characters represent forward sortation areas codes had significant limitations. The partial codes changed (FSAs), defined by Canada Post to sort mail into coarse geo- the boundaries of service areas and were thus unreliable de- graphic areas to speed up mail delivery. As of January 1988, scriptions of core population groups within those service ar- there were 1296 FSAs across Canada, 99 of which were in Met- eas. Because they also shifted geographic locations, some- ropolitan Toronto (9). The geographic coordinates for a three- times up to a kilometre, they also raised questions about the digit FSA's position are at the geometric centroid of a FSA, es- allocation of submitting locations.
sentially an arbitrary point without any reference to the phe- nomenon being mapped.
DATA AND METHODS
The complete six-digit postal code, which includes the FSA The data, consisting of 1329 isolates submitted for the 133 and another three characters, denotes local delivery units, submitting locations in Metropolitan Toronto between 1988 typically city blocks. These postal codes are associated with and 1992, were provided by the National Laboratory for Sexu- points on the map known as block face centroids. A block face Can J Infect Dis Vol 8 No 5 September/October 1997 G: INFDIS 1997 Vol8no5 decker.vp Thu Oct 02 15:18:05 1997 Color profile: Disabled Composite Default screen Mapping antibiotic-resistant N gonorrhoeae isolates
is one side of a city street, and its centre is equidistant be- tween consecutive intersections with other streets. There are 48,136 block face centroids in Metropolitan Toronto. The cen- troid is a good estimate (within a city block) of the submittinglocation, yet is still just an estimate. This is important to un-derstand when questions of confidentiality are raised.
The number of isolates sent by a submitting location was partly a function of the nature of the facility. For example, 43 of 133 submitting locations reported sending only one isolate.
Medical diagnostic laboratories and special STD clinics sub-mitted higher numbers of isolates. One location submitted148 isolates, while the average submitted across all 133 loca-tions was 10 isolates (Figure 1). Because of their exclusivemandate, STD clinics were differentiated from all other sub-mitting locations.
Figure 1) Number of submitting locations in Metropolitan Toronto,
Figure 2 shows the geographical distribution of antibiotic- 1988 to 1992, by frequency of isolates reported resistant N gonorrhoeae isolates across submitting locationsin Metropolitan Toronto. Figure 3 shows the distribution of ting locations were confined within the one FSA, giving the isolates at the more detailed scale of downtown Toronto. Fig- impression that only one FSA was serviced. This is not the ures 2 (top) and 3 (top) represent submitting locations of case on Figure 2 (bottom), where several of the previously ag- N gonorrhoeae isolates at the three-digit FSA level. Figures 2 gregated locations were more accurately positioned on the (bottom) and 3 (bottom) used the complete six-digit postal boundary between two FSAs. Point 2 on the maps demon- code for submitting locations of N gonorrhoeae isolates.
strated another similar example. Point 3 on the maps illus- Regardless of the type of submitting location, the overall trates a striking example of the total displacement of a STD pattern revealed by Figure 2 (top) is hierarchical. Downtown clinic. In Figure 2 (top), point 3 is located in the centre of an Toronto had the highest concentration of submitting locations, FSA, whereas in Figure 2 (bottom) this same clinic is associ- several with high isolate counts. Surrounding the downtown ated with three different FSAs.
area was a semicircular ring, in which most submitting loca- When analyzing the information on Figure 3 (top and tions reported low or moderate isolate counts. Submitting loca- bottom) on a larger scale, issues of positional accuracy tions with high isolate counts were located in the northern, became even more apparent. The submitting location report- eastern and western suburbs of Metropolitan Toronto.
ing the largest number of isolates in the upper FSA moved at STD clinics handled less than one-quarter of the total least 1 km to the southeast of that FSA on Figure 3 (bottom) number of antibiotic-resistant isolates submitted. Of the 133 when six-digit postal codes were used. The location of the STD submitting locations, 19 were STD clinics, accounting for 237 clinics shifted dramatically from a random pattern in Figure of the 1321 isolates. Most of the STD clinics were concentrated 3 (top) to a clustered pattern around Yonge Street in Figure 3 downtown. Of the few STD clinics in the suburbs, only the (bottom). Figures 2 and 3 demonstrate the point that the spa- Scarborough clinic on the east side submitted a moderately tial patterns of submitted antibiotic-resistant N gonorrhoeae high isolate count. In contrast, three STD clinics in the centre vary considerably when working at the two levels of aggrega- of the map, east and west of Yonge Street submitted only a few tion possible with the Canadian postal code data.
A comparison of the two maps in Figure 2 demonstrates that the use of FSA postal codes compared with complete six- Epidemiological studies of different infectious diseases digit postal codes. Forty-four submitting locations were lo- have noted the importance of working at multiple scales cated at any one of the 99 FSA centroids in Metropolitan To- (10,11). Three-digit postal codes are generally effective in ronto (Figure 2 [top]), mapped using the partial codes. In summarizing overall geographical distributions when mapped Figure 2 (bottom), 75 submitting locations were located at any at the metropolitan level. However, when mapped at a neigh- one of the possible 48,136 block face centroids. The complete bourhood scale, the use of three-digit postal codes reduces the postal code data revealed more pronounced clustering of sub- number of submitting locations and visibly reduces their posi- mitting locations, especially evident in downtown Toronto tional accuracy. The movement of submitting locations away and throughout the suburbs (Figure 2 [bottom]).
from their true locations, sometimes as much as a kilometre, The necessity for increased positional accuracy also be- can have considerable implications. Inaccuracies in the posi- came apparent when delineating arbitrary service (catchment) tion of submitting locations may lead to misidentification of areas for submitting locations. Consider, for example, the core groups in the transmission of N gonorrhoeae. Descrip- service area for submitting locations at designated point 1 on tions of the socio-economic characteristics of a population liv- Figure 2 (top). At the three-digit postal code level, four submit- ing near a medical clinic involves defining a service area Can J Infect Dis Vol 8 No 5 September/October 1997 G: INFDIS 1997 Vol8no5 decker.vp Thu Oct 02 15:18:09 1997 Color profile: Disabled Composite Default screen Decker et al
Figure 2) Top Spatial distribution of penicillinase-producing isolates of Neisseria gonorrhoeae (PPNG) and tetracycline-resistant N gonorrhoeae
(TRNG) by submitting locations in Metropolitan Toronto, 1988 to 1992, using three-digit postal codes. Bottom Spatial distribution of PPNG and
TRNG by submitting locations in Metropolitan Toronto, 1988 to 1992, using six-digit postal codes. Symbols (circles and boxes) are proportionally
graduated according to the number of isolates reported at each submitting location. Forward sortation area boundaries are included as a frame ofreference. STD Sexually transmitted disease around the clinic. Census data, which are associated with the considerably. These shifting patterns also raise questions area falling within a service area, are then summed. Depend- about the efficiency of the current geographical allocation of ing on the accurate location of clinics and variations in the size and shape of their service areas, the socio-economic char- STDs can be usefully analyzed on both a city and neigh- acteristics of the core transmission group described can differ bourhood scale. If the goal of such an analysis is to derive an Can J Infect Dis Vol 8 No 5 September/October 1997 G: INFDIS 1997 Vol8no5 decker.vp Thu Oct 02 15:18:15 1997 Color profile: Disabled Composite Default screen Mapping antibiotic-resistant N gonorrhoeae isolates
overall sense of the geographic pattern of N gonorrhoeae re- porting throughout a large metropolitan region or the ration- alization of services and allocation of scarce resources, or to answer questions about where to place a new facility, then theuse of three-digit postal codes is appropriate. Differences be-tween suburbs and the central city, or between public health unit jurisdictions within the metropolitan region can be read- ily discerned. Further analysis of the generalized pattern might relate the distribution of STD submitting locations to the more generalized socio-demographic patterns and trans- portation infrastructure of the city.
At the neighbourhood scale, on the other hand, if the re- search goal is to analyze the behavioural patterns of clients orto associate submitting locations with population characteris-tics, the full postal code is more appropriate. Use of clientpostal code data has produced several important results. Astudy by Potterat et al (12) in Colorado in 1985 demonstratedthat geographic clustering of STDs was evident at the censustract level within downtown areas. Rothenberg (13) analyzedall submitted cases of N gonorrhoeae (resistant and nonresis-tant strains) in upstate New York, using data combined from1975 to 1980. He was able to identify high prevalence censustracts that he suggested may be responsible for continuing en-demicity of the disease in that state. Studies of N gonorrhoeaein other places have also shown that further geographicalanalysis can identify core transmission groups and locate coretransmission areas (14,15).
Ecological studies on social and sexual networks and sup- port systems have shown that STDs tend to concentrate inneighbourhoods (16,17). In Canada, only one ecological study Figure 3) Top Spatial distribution of penicillinase-producing isolates
of STDs has been done, yet it relied on three-digit postal codes of Neisseria gonorrhoeae (PPNG) and tetracycline-resistant N gonor- to detect and compare high rate areas with census tract char- rhoeae (TRNG) by submitting locations in Toronto, 1988 to 1992, us- acteristics within those areas. This report found that N gonor- ing three-digit postal codes. Bottom Spatial distribution of PPNG and
rhoeae outbreaks occur in "low-income minority populations TRNG by submitting locations in Toronto, 1988 to 1992, using six-digit postal codes. Symbols (circles and boxes) are proportionally involving drug use and high risk sexual behaviour" (18).
graduated according to the number of isolates reported at each submit- The use of six-digit postal codes in the above Canadian ting location. Forward sortation area boundaries are included as a study would have been more accurate. Confidentiality would frame of reference. STD Sexually transmitted disease not have been compromised because six-digit codes are onlylocational estimates centred on one side of a street of a cityblock. Furthermore, when mapped, ‘confidential' data are of- ten aggregated by several years in order to get an adequate Depending on the research question, careful consideration number of cases and then grouped into ranges. This chal- must be given to the use of partial versus complete postal lenges the reticence and rigidity of some institutions, agencies codes in analyzing data. We used submitting location data to and government departments that restrict data accessibility demonstrate this methodological fact. An analysis that related on the basis of confidentiality. Moore (19) referred to such re- submitting location data to place of residence data would be strictions as the "one-legged Tarzan syndrome". He asserted very valuable and it is clear from the discussion that confiden- that it is indeed highly improbable that an individual could tiality would not be compromised. Questions could then be ad- ever be recognized, let alone any inferences made, from link- dressed such as: What type of facility do clients favour given ages between aggregated cases and socio-economic data.
they have a choice in their immediate vicinity? How far are Moore's thoughts are echoed by another prominent geogra- they willing to travel to receive treatment? What is the per- pher, Peter Gould (20), who models the geographic spread of centage of repeat cases (not isolates) of individuals out of the diseases. Gould stated that in all studies of mathematical total number of submissions? To move on to these more com- modelling, geographers have never identified any one person, plicated issues of the geography of STDs, it is important to un- have no need to do so and would not know what to do with the derstand the ramifications of using three- versus six-digit information if they had it. He added that the loss of confidenti- postal codes, regardless of the type of data. However, in issues ality "is a genuine fear, but it has been taken to extreme and of such sensitivity, as with STDs, positional accuracy becomes absurd lengths" (20).
all the more important.
Can J Infect Dis Vol 8 No 5 September/October 1997 G: INFDIS 1997 Vol8no5 decker.vp Thu Oct 02 15:18:22 1997 Color profile: Disabled Composite Default screen Decker et al
gonorrhoeae in Ontario: Trends 1989-1990. Can Commun Dis ACKNOWLEDGEMENTS: We thank Dr KH Yeung (NLSTD) for his as-
Rep 1992;18:32-4.
sistance in downloading data; Dr C Krishnan, Central Public Health 9. Geographic Division, Statistics Canada. Postal code conversion Laboratory, Laboratory Services Branch, Ontario Ministry of Health, file documentation. Ottawa: Statistics Canada, 1988.
Etobicoke, Ontario, for sharing isolate information with the NLSTD; 10. Schneider D, Greenberg MR, Donaldson MH, Choi D. Cancer Grant Simpson for data manipulation and mapping; and Melinda clusters: The importance of monitoring multiple geographic Hecht-Enns for the final maps.
scales. Soc Sci Med 1993;37:753-9.
11. Thomas R. Geomedical Systems. Intervention and Control.
London: Routledge, 1992:159-92.
12. Potterat JJ, Rothenberg RB, Woodhouse DT, Muth JB, Pratts CI, 1. Minister of Health Canada. Notifiable Diseases Annual Summary, Fogle JS. Gonorrhea as a social disease. Sex Trans Dis 1993. Can Commun Dis Rep 1995;21(Suppl 1):24.
2. Yeung KH, Dillon JR, and the National Study Group. Cases of 13. Rothenberg RB. The geography of gonorrhea, empirical Neisseria gonorrhoeae with plasmid-mediated resistance to demonstration of core group transmission. Am J Epidemiol penicillin increase. Can Dis Wkly Rep 1990;16:13-6.
3. Yeung KH, Pauze M, Dillon JR, and the National Study Group.
14. May RM. The transmission and control of gonorrhea. Nature Status of penicillinase-producing Neisseria gonorrhoeae in Canada – 1989. Can Dis Wkly Rep 1991;17:49.
15. Rice R, Roberts PL, Handsfield H, Holmes KK. Sociodemographic 4. Brown S, Terro R, Riley G, Harnett N, Krishnan C. Increasing distribution of gonorrhea incidence: Implications for prevention resistance to antimicrobial agents among isolates of Neisseria and behavioral research. Am J Public Health gonorrhoeae in Ontario: Trends 1989-1990. Can Comm Dis Rep 16. Greenbaum SD, Greenbaum PE. The ecology of social networks in 5. Dillon JR, Carballo M. Molecular epidemiology and novel four urban neighbourhoods. Soc Networks 1985;27:47-76.
combinations of auxotype, serovar, and plasmid content in 17. Hammer M. ‘Core' and ‘extended' social networks in relation to tetracycline-resistant Neisseria gonorrhoeae isolated in Canada.
health and illness. Soc Sci Med 1983;17:467-70.
Can J Microbiol 1990;36:64-7.
18. Shahin R, Wallace E. Gonorrhea – North York, 1990. Public 6. Harnett N, Brown S, Riley G, Terro R, Krishnan C. Decreased Health Epidemiol Rep Ontario 1991;2:277-9.
susceptibility of Neisseria gonorrhoeae to fluoroquinolones – 19. Moore EG. Research agendas and Statistics Canada products: Ontario, 1992-1994. Can Comm Dis Rep 1995;21:17-20.
geographical perspectives on micro-data files. Operational 7. Yeung KH, Dillon JR. Cases of Neisseria gonorrhoeae with plasmid-mediated resistance to penicillin increase. Can Dis Wkly 20. Gould P. Modelling the geographic spread of AIDS for Rep 1990;16:13-6.
educational intervention. In: R Ulack, WF Skinner, eds. AIDS 8. Brown S, Terro R, Riley G, Harnett N, Krishnan C. Increasing and the Social Sciences. Lexington: The University Press of resistance to antimicrobial agents among isolates of Neisseria Kentucky, 1991:31.
Can J Infect Dis Vol 8 No 5 September/October 1997 G: INFDIS 1997 Vol8no5 decker.vp Thu Oct 02 15:18:23 1997

Source: http://scholars.wlu.ca/cgi/viewcontent.cgi?article=1016&context=geog_faculty