Printer Friendly

Examining the association between neighbourhood characteristics and gonorrhea rates among women aged 15 to 24 years in Montreal, Canada.

Gonorrhea has been nationally notifiable in Canada since 1924 and remains the second most commonly reported sexually transmitted infection (STI) in the country after chlamydia. (1) In Canada, there has been a gradual but steady increase in reported rates of gonorrhea in both sexes since 2000. The reported incidence has increased from 20.1/100,000 in 2000 to 29.6/100,000 in 2010. The incidence rate of gonorrhea for Montreal is higher than that for both Canada (with the exception of 2004) and Quebec. From 2000 to 2010, the incidence rate of reported cases in Montreal doubled from 24.5 to 51.5 per 100,000.

The increase in the reported incidence of gonorrhea in Montreal has affected both men and women and all age groups; however, males continue to have higher incidence rates compared to females (2010: 78/100,000 for males and 26/100,000 for females). Although the highest incidence rate of gonorrhea is consistently among men, from 2005 to 2010, the percent increase in the incidence rate was almost five times higher among women (from 8/100,000 to 26/100,000: +225%) than among men (from 53/100,000 to 78/100,000: +47%). In women, the increase was particularly high in the 15-19 year and 20-24 year age groups (see Figure 1).

Because of the greater risk of severe complications in females, the Montreal Health and Social Services Agency has focused their investigations on women under the age of 25 years. However, the reasons for the increase among young women are not fully understood. Therefore, the objective of this study was to determine whether gonorrhea incidence rates among young women were associated with neighbourhood-level population characteristics, in order to help target intervention strategies.

Numerous studies have identified individual-level risk factors for STIs and as a result, public health typically targets high-risk individuals, focusing on screening, treatment, partner notification and counselling. (2,3) However, using individual case characteristics is not always sufficient to fully understand surveillance trends or for the control of outbreaks. Individual characteristics do not always adequately describe key groups, making it difficult to design effective interventions to prevent gonorrhea transmission. As a result, there has been increased attention paid to the social determinants of disease and the relationship between gonorrhea rates and population characteristics. These geographic approaches to STI prevention in areas where core groups exist have been suggested as feasible alternatives to case-by-case management and have had success in lowering incidence. (4) A more thorough understanding of the target population combined with geographic information may help refine current uniformly applied intervention strategies or with the development of new targeted strategies.

[FIGURE 1 OMITTED]

METHODS

Incident gonorrhea cases were defined as female residents of Montreal, aged 15 to 24 years, who met Quebec's provincial gonorrhea surveillance definition, (5) with a notification date from 2002 to 2009. A person may be included multiple times during the study period if repeated infections (as opposed to relapses) were reported.

Neighbourhoods were used as the analysis unit to approximate the concept of community. These 111 non-administrative boundaries were defined in 2007 by the Montreal Health and Social Services Agency in consultation with many local partners. Neighbourhood populations range from 1,975 to 64,100; with an average population of 16,706. To define the neighbourhoods, the Montreal Health and Social Services Agency conducted a review and analysis of all documents and maps developed by the city and other local territories. This included neighbourhood profiles describing in detail the characteristics of the population, documents outlining the history of the neighbourhood, or any other publication reporting a division of the territory. The purpose was to divide the health and social territories in the Montreal area into smaller units that reflect distinct social realities and, in many cases, areas requiring different interventions. Each neighbourhood had to be fully included in a local community service centre. Specifically, the boundaries of neighbourhoods had to be drawn from census dissemination areas and to follow the perimeter of the local community service centre. A neighbourhood therefore would consist of a set of dissemination areas or parts thereof. Consultations were held to obtain consensus on the defined neighbourhoods. Participants included social workers and nurses, and representatives from community organizations, day care centres, schools, the Ministry of Immigration and Cultural Community, the city of Montreal and the Ministry of Families and Seniors.

The smaller units were chosen because the compilation of data by larger geographical boundaries can mask important differences. (6) Neighbourhood characteristics were obtained from 2006 Canadian Census data.

The dependent variable was the neighbourhood gonorrhea incidence rate for females aged 15 to 24 years. The independent variables included material and social deprivation indices, their combination and components, and ethnic origin (see Table 1). These variables were selected based on the deprivation index created in Quebec as well as on findings from previous research with specific interest in racial/ethnic composition and STI incidence. (7-10)

Adjusted incidence rate ratios (IRR) were estimated by negative binomial regression to reflect the expected changes in the gonorrhea rate associated with changes in the neighbourhood characteristics.

Variables were applied one at a time and as a group (deprivation indices) to determine whether a single variable or all variables combined were associated with disease rates. Variables significant at p<0.05 in the univariate analysis were further examined in the multivariate models. In the final model, independent variables were normalized into z scores to facilitate comparison of their respective IRRs. STATA version 10.1 was used for analyses. This work only used publicly available aggregate data and no ethical review was required.

RESULTS

From 2002 to 2009, 837 gonorrhea cases were reported among females aged 15 to 24 years. Of those cases, 784 (93.7%) could be geographically coded and attributed to a neighbourhood. Twenty neighbourhoods throughout the island of Montreal had cumulative incidence rates of over 130 per 100,000 population (Figure 2). Less than half (8/20: 40%) of these neighbourhoods were among those with greater than 50% of the population who are both materially and socially deprived. Therefore, the remaining 12 neighbourhoods with high incidence rates of gonorrhea were among those with more favourable material and social conditions.

Results from the univariate analyses are shown in Table 2. Gonorrhea incidence was significantly associated with average income after taxes, and the proportion of the population: without high school diploma; in single-parent families; separated, widowed or divorced; materially deprived; socially and materially deprived; or whose origin is Caribbean, Aboriginal, Western European, Northern European, Eastern European, and African (Table 2).

The six independent variables and six ethnic origin groups that were significantly associated with gonorrhea rates were included in the multivariate analysis. The first multivariate model revealed that when the deprivation indices were included with some of their components, they were no longer associated with the rates of gonorrhea (Material: IRR 1.004, p=0.290, CI 0.996-1.012; Material and Social: 1.000, p=0.984, CI 0.990-1.009). Only marital status, no diploma and three ethnic origin groups (Aboriginal, Caribbean, and African) remained associated. However, in subsequent multivariate models, "no diploma" and "marital status" (no diploma: IRR 0.99, p=0.215, CI 0.97-1.01; Marital Status: IRR 1.011, p=0.170, CI 0.991.03) ceased to be associated. The final model therefore only included African, Aboriginal and Caribbean ethnic origin groups.

[FIGURE 2 OMITTED]

The final analysis, using normalized independent variables, revealed that higher proportions of African, Aboriginal and Caribbean populations were associated with rates of gonorrhea, even after controlling for indices of deprivation (see Table 3). African origin population had the strongest association, with each increase of one standard deviation in the percentage of the population whose origin is African being associated with a 34% increase in the rate of gonorrhea among women aged 15 to 24 years.

Negative binomial regression was chosen instead of Poisson regression because the variance for the independent variable was five times greater than the mean (mean 7.06 and variance 36.8), meaning that the data were overdispersed and the Poisson regression method therefore was not appropriate.

DISCUSSION

This study demonstrates the application of regression models to examine the association between neighbourhood-level characteristics and reported incidence rates of gonorrhea. Our results imply that each increase of one standard deviation in the proportion of the population whose origin is African, Aboriginal or Caribbean is associated with a 34%, 33% or 18% increase, respectively, in the rate of gonorrhea among women aged 15 to 24 years. These results suggest that the proportions of three ethnic origin groups are either markers for risk factors of gonorrhea or could be themselves risk factors. The reasons for the association between gonorrhea incidence and these ethnic origin groups are difficult to identify. Late seeking of health care services and failure to use condoms do not seem to account for the association. (7,13) As well, there are no known biological susceptibility differences between ethnic groups. (14)

Sexual mixing patterns have been suggested as a possible mechanism for increased transmission. Some have suggested that sexually transmitted infections stay within certain ethnic groups because partner choices are more segregated (assortative mixing) than in other groups. (15,16) Rothenberg et al. demonstrate that in groups with high STI risk, partner concurrency is higher, even though the number of partners over time may be similar. (3)

Previous studies have shown that STI rates are highest in areas with lower levels of socio-economic status; (8,10,17,18) however, our results did not demonstrate that among young women in Montreal an association exists between gonorrhea rates and traditional socioeconomic status indicators associated with STIs, such as unemployment or material and social deprivation indicators.

We note limitations in the gonorrhea incidence data that were used. They were obtained from reported cases of infection, which are dependent upon medical practitioners testing for infections and on physicians or laboratories reporting positive results. Variations in STI screening practices among physicians may have influenced the number of reported cases.

By focusing only on neighbourhoods of residence of cases, our analysis ignored spatial effects from contiguous neighbourhoods. This omission may have affected our estimates, but it is difficult to determine the extent of the effect. Spatial smoothing is one method that could have been used; however, consensus on appropriate methods for adjusting STI rates to increase reliability of comparisons has not been reached. (19,20)

STI control strategies currently focus on individual intervention. (21-23) In the short term, these strategies may have an impact on interrupting disease transmission. (24) However, it has become increasingly apparent that risk factors for STIs cannot be explained solely by individual behaviour. (25,26) This study is consistent with current findings that suggest that differences in population characteristics (more specifically, ethnicity) are associated with differences in gonorrhea rates. (7-10,19,25-30) Although ethnicity may be strongly correlated with socio-economic status, our findings demonstrate that there are factors of ethnicity that transcend poverty. (29) Our study adds to the evidence that suggests that in order to achieve and sustain reductions in community-level disparities of STIs, public health organizations must pay greater attention to the population characteristics. Inclusion of population data into existing surveillance systems will facilitate the study of their relationship to disease trends at the community level.

Community interventions are rare in the field of prevention and control of STIs, as traditional control strategies have focused primarily on individual characteristics of cases to determine risk factors. However, this study demonstrates that gonorrhea is clustered in neighbourhoods that have high proportions of African, Aboriginal and Caribbean populations. In small cities, it may be possible for the local public health department to interview and follow up on every case to obtain individual characteristics. However, disease rates in Montreal have become and will remain a problem insurmountable with current approaches. Population-level prevention programs could be a useful addition to the current armamentarium.

Conflict of Interest: None to declare.

REFERENCES

(1.) Public Health Agency of Canada. Brief Report on Sexually Transmitted Infections in Canada: 2007. Ottawa, ON: PHAC, 2009.

(2.) Centers for Disease Control and Prevention. Program operations guidelines for STD prevention. Atlanta, GA: CDC, 2001.

(3.) Rothenberg R, Kimbrough L, Lewis-Hardy R, Heath B, Williams OC, Tambe P, et al. Social network methods for endemic foci of syphilis: A pilot project. Sex Transm Dis 2000;27(1):12-18.

(4.) Zenilman J, Ellish N, Fresia A. The geography of sexual partnerships in Baltimore: Applications of core theory dynamics using a geographic information system. Sex Transm Dis 1999;26:75.

(5.) La Direction des communications du ministere de la Sante et des Services sociaux du Quebec. Definitions nosologiques: Maladies d'origine infectieuse, 7e edition. 2008.

(6.) Direction de sante publique, Agence de la sante et des services sociaux de Montreal. Principales caracteristiques de la population de la region de Montreal. 2009;1.

(7.) Barry P, Kent C, Klausner J. Risk factors for gonorrhea among heterosexuals-San Francisco, 2006. Sex Transm Dis 2009;36(2):S62.

(8.) Du P, McNutt L, O'Campo P, Coles F. Changes in community socioeconomic status and racial distribution associated with gonorrhea rates: An analysis at the community level. Sex Transm Dis 2009;36(7):430.

(9.) Low N, Sterne J, Barlow D. Inequalities in rates of gonorrhoea and chlamydia between black ethnic groups in south east London: Cross sectional study. Sex Transm Dis 2001;77:15.

(10.) Thomas J, Gaffield M. Social structure, race, and gonorrhea rates in the southeastern United States. Ethnicity Dis 2003;13:362.

(11.) Direction de sante publique, Agence de la sante et des services sociaux de Montreal. Regard sur la defavorisation a Montreal. 2008;1.

(12.) Pampalon R, Raymond G. Un indice de defavorisation pour la planification de la sante et du bien-etre au Quebec. Maladies chroniques au Canada 2000;21(3):113.

(13.) Kramer MA, Aral SO, Curran JW. Self-reported behavior patterns of patients attending a sexually transmitted disease clinic. Am J Public Health 1980;70(9):997-1000.

(14.) Stevens-Simon C, Jamison J, McGregor JA, Douglas JM. Racial variation in vaginal pH among healthy sexually active adolescents. Sex Transm Dis 1994;21(3):168-72.

(15.) Laumann EO, Youm Y. Racial/ethnic group differences in the prevalence of sexually transmitted diseases in the United States: A network explanation. Sex Transm Dis 1999;26(5):250-61.

(16.) Fenton K, Johnson AM, Nicoll A. Race, ethnicity, and sexual health. BMJ1997;314(7096):1703-4.

(17.) Rice R, Roberts P, Handsfield H, Holmes K. Sociodemographic distribution of gonorrhea incidence: Implications for prevention and behavioural research. Am J Public Health 1991;81(10):1252.

(18.) Cohen D, Spear S, Scribner R, Kissinger P, Mason K, Wildgen J. "Broken Windows" and the risk of gonorrhea. Am J Public Health 2000;90(2):230-36.

(19.) Owusu-Edusei K, Chesson H. Using spatial regression methods to examine the association between count-level racial/ethnic composition and resported cases of chlamydia and gonorrhea: An illustration with data from the state of Texas. Sex Transm Dis 2009;36(10):657.

(20.) Waller L, Gotway C. Applied Spatial Statistics for Public Health Data. Hoboken, NJ: John Wiley and Sons, 2004.

(21.) Metzler M. Social determinants of health: What, how, why, and now. Prev Chronic Dis 2007;4(4):A85.

(22.) Syme S. Social and economic disparities in health: Thoughts about intervention. Milbank Q 1998;16:493.

(23.) Syme S. Social determinants of disease. Ann Clin Res 1987;19(2):44-52.

(24.) Low N, FitzGerald MR. Success and failure in gonorrhea control. Dermatol Clin 1998;16(4):713-22.

(25.) Hallfors DD, Iritani BJ, Miller WC, Bauer DJ. Sexual and drug behavior patterns and HIV and STD racial disparities: The need for new directions. Am J Public Health 2007;97(1):125-32.

(26.) Ellen J, Aral S, Madger L. Do differences in sexual behaviours account for the racial/ethnic differences in adolescents' self-reported history of a sexually transmitted disease? Sex Transm Dis 1998;25(3):125-29.

(27.) Becker KM, Glass GE, Brathwaite W, Zenilman JM. Geographic epidemiology of gonorrhea in Baltimore, Maryland, using a geographic information system. Am J Epidemiol 1998;147(7):709-16.

(28.) Ellen JM, Hessol NA, Kohn RP, Bolan GA. An investigation of geographic clustering of repeat cases of gonorrhea and chlamydial infection in San Francisco, 1989-1993: Evidence for core groups. J Infect Dis 1997;175(6):1519-22.

(29.) Krieger N. Refiguring "race": Epidemiology, racialized biology, and biological expressions of race relations. Int J Health Serv 2000;30(1):211-16.

(30.) Lacey CJ, Merrick DW, Bensley DC, Fairley I. Analysis of the sociodemography of gonorrhoea in Leeds, 1989-93. BMJ 1997;314(7096):1715-18.

Received: April 19, 2012

Accepted: July 18, 2012

Nashira J. Khalil, MHSA, [1,2] Robert Allard, MD [2,3]

Author Affiliations

[1.] Canadian Field Epidemiology Program, Public Health Agency of Canada, Ottawa, ON

[2.] Montreal Health and Social Services Agency, Montreal, QC

[3.] Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, QC

Correspondence: Nashira Khalil, Public Health Agency of Canada, 200 Boulevard Rene Levesque West, Montreal, QC H2Z 1X4, Tel: 514-496-7904, Fax: 514-2833309, E-mail: nashira.khalil@phac-aspc.gc.ca
Table 1. Independent Variables

Independent Variables

1. Population aged 15 years and older without a diploma

2. Employment rate for the population aged 15 years or older

3. Average income after taxes

4. Single-parent families among families with children

5. Population living alone

6. Population separated, divorced, widowed

7. Ethnocultural group (French, Caribbean, British Isles, Aboriginal,
Other North American, Latin American, Central and South American,
Western European, Northern European, Eastern European, Southern
European, African, Arab, West Asian, South Asian, East or South East
Asian, Oceania, Maghreb)

8. Material deprivation which reflects the deprivation of goods and
conveniences of everyday life (% of the population without a diploma;
employment rate; average income) (11)

9. Social deprivation which reflects the fragility of the social
network and family to the community (% of the population separated,
divorced, widowed; % of single-parent families; % of the population
living alone (11)

10. Material and social deprivation: Characterizes a state of relative
disadvantage of individuals, families or groups with respect to a
population to which they belong: a local community, region or nation.
Includes the indicators that make up both social and material
deprivation (11,12)

Table 2. Results From Univariate Negative Binomial Regression:
Unadjusted Incidence Rate Ratio (IRR) Associated With a Unit
Increase in Each Independent Variable

Independent Variable                   IRR     p-value      95% CI

% of the population aged 15 years     1.027#   0.000#    1.012-1.042#
  and older without a high school
  diploma#
Employment rate for the population    0.986     0.231    0.964-1.008
  aged 15 years or older
Average income after taxes#           0.999#   0.000#    0.999-0.999#
% of single-parent families among     1.033#   0.000#    1.020-1.047#
  families with children#
% of population living alone          1.004     0.608    0.986-1.024
% of population separated,            0.981#   0.020#    0.966-0.997#
  divorced, widowed#
% of population socially deprived     0.997     0.525    0.990-1.004
% of population materially            1.005     0.054    0.999-1.011
  deprived#
% of population socially and          1.013     0.000    1.006-1.019
  materially deprived#
% of population whose origin          1.001     0.851    0.986-1.016
  is French
% of population whose origin          1.059#   0.000#    1.028-1.092#
  is Caribbean#
% of population whose origin is       0.987     0.140    0.971-1.004
  British Isles
% of population whose origin is       1.157#   0.003#    1.049-1.276#
  Aboriginal#
% of population whose origin is       1.004     0.360    0.995-1.012
  Other North American
% of population whose origin is       1.057     0.114    0.986-1.133
  Latin American, Central or
  South American
% of population whose origin is       0.927#   0.006#    0.879-0.978#
  Western European#
% of population whose origin is       0.732#   0.021#    0.562-0.954#
  Northern European#
% of population whose origin is       0.969#   0.017#    0.945-0.994#
  Eastern European#
% of population whose origin is       0.991     0.248     0.977-1.00
  Southern European
% of population whose origin is       1.190#   0.000#    1.120-1.260#
  African#
% of population whose origin is       0.985     0.210    0.964-1.000
  Arab
% of population whose origin is       0.967     0.333    0.905-1.030
  West Asian
% of population whose origin is       1.011     0.462    0.980-1.044
  South Asian
% of population whose origin is       1.000     0.987    0.974-1.027
  East or South East Asian
% of population whose origin is       0.572     0.482    0.120-2.717
  Oceania
% of population whose origin is       0.941     0.091    0.877-1.009
  Maghreb

Bolding in the table represents the independent variables that were
significantly associated with gonorrhea rates in the multivariate
analysis.

Note: Bolded characters are indicated with #.

Table 3. Final Model With Normalized Data: Adjusted Incidence Rate
Ratio (IRR) Associated With a 1 Standard Deviation Increase in Each
Independent Variable

Independent Variable         IRR for an Increase      Same IRR,
                                of 1 Standard        Adjusted for
                                Deviation IRR        Deprivation
                                  (95% CI)           IRR (95% CI)

% of the population           1.34 (1.20-1.49)     1.35 (1.21-1.52)
whose origin is African
% of the population           1.32 (1.19-1.46)     1.36 (1.18-1.57)
whose origin is Aboriginal
% of the population           1.19 (1.07-1.33)     1.20 (1.08-1.34)
whose origin is Caribbean
% of the population both              -                 0.998
materially and socially                             (0.990-1.004)
deprived
COPYRIGHT 2012 Canadian Public Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:QUANTITATIVE RESEARCH
Author:Khalil, Nashira J.; Allard, Robert
Publication:Canadian Journal of Public Health
Article Type:Report
Geographic Code:1CANA
Date:Sep 1, 2012
Words:3546
Previous Article:A spatial analysis of asthma prevalence in Ontario.
Next Article:The need for economic evaluation in primary prevention of cancer.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters