Factors associated with pregnancy and STI among aboriginal students in British Columbia.
Aboriginal young people in Canada and the US are over-represented in adolescent pregnancy and sexually transmitted infection (STI) statistics and under-represented in sexual health research. (1-3) The reasons for disparities in sexual health outcomes are not fully understood, and indigenous scholars stress the importance of colonial history, lack of access to health care services and socio-economic disadvantage. (4,5) Even less is known about how to effectively intervene to reduce sexual risk.
Observational literature that could inform intervention development for this population is scant. The few studies that focus on the sexual health of Aboriginal young people tend to be qualitative in nature or use small samples from particular Nations. (6,7) Only one large US study examining correlates of adolescent pregnancy in a school-based sample of indigenous adolescents has been conducted, which looked at pregnancy and sexual orientation. (8) One population-based survey conducted in Canada during the 1990s focused on STI, but it sampled adults living on-reserve in Ontario. (9) Several existing cohort studies in BC do include Aboriginal young people, but are restricted to injection drug and other substance users in Vancouver and/or Prince George. (10,11) Thus there is a clear need for information on factors underlying increased risk of adolescent pregnancy involvement and STI diagnosis among broader populations of young Aboriginal people in Canada.
The present study sought to determine the key predictors of self-reported pregnancy involvement and STI diagnosis in a large probability sample of Aboriginal secondary school students.
Secondary analyses were conducted using data from the British Columbia Adolescent Health Survey 2003 (BC AHS 2003). (12) The AHS invites all 59 BC School Districts to participate, and in 2003, 45 agreed to allow access to students. In these 45 districts, a stratified random sample of 1,557 grade 7-12 classrooms from all types of schools (including those on- and off-reserve) was selected. All 40,040 students from these classrooms were eligible to participate. Informed consent was sought either from students (with parental notification) or parents (with student assent). The response rate among students was 76%, or 30,588 usable questionnaires in the participating districts. Students who did not participate were absent on the day of the survey (12%), failed to return the parental consent form (7%), or they/their parents refused to participate (3%). More details about survey design are available elsewhere. (12)
Ethical consultation and approval
At the outset of the project, Aboriginal community members were approached to ensure the proposed work would be conducted in an ethical and sensitive way that would directly benefit the Aboriginal community. Subsequently, institutional approvals were granted from the London School of Hygiene and Tropical Medicine and University of British Columbia Ethical Review Boards.
Only young people who self-identified as Aboriginal/First Nations and answered yes to the question "Have you ever had sexual intercourse ("gone all the way")?" were included in analyses.
Measures in the BC AHS relevant to these analyses were mainly drawn from large-scale school-based surveys from the US. (13,14) Both outcomes--self-reports of ever having been pregnant and ever having been diagnosed with an STI by a doctor or other health professional--were measured using binary variables. Eleven exposure variables and one confounder (age) were examined in relation to each outcome (Table 1). Scores for family and school connectedness and substance use measures were computed by summing item scores and dividing by the number of items completed. School and family connectedness were modeled as continuous variables. Level of substance use for individuals were classified as either above or below the median lifetime substance use for all Aboriginal young people, and modeled as a binary variable. Cronbach's alpha for school connectedness was 0.81 for young women and 0.80 for young men; for family connectedness, alpha was 0.86 for both sexes.
All analyses were conducted using STATA 9.0, (15) which handles weighted data and correctly estimates variance while accounting for complex survey sampling. Logistic regression was used to calculate unadjusted and adjusted odds ratios for each association. Backward elimination was used to fit a multivariate logistic regression model to determine the most important predictors of pregnancy and STI diagnosis. All variables were entered into the model, and the one with the lowest odds ratio and p-value was removed. The model was re-run and this step repeated until only variables with p<0.05 remained.
Participants with missing data on particular items were excluded from analyses involving those items. Outcomes were not related to missing one or more responses (p>0.05 for each).
In the BC AHS 2003, 2,476 participants identified as Aboriginal (8.1% of the full AHS sample). Of 1,336 young women, 445 (34.8%) had ever had sex. Of 1,140 young Aboriginal men, 360 (33.7%) had ever had sex. Among young women who had ever had sex, 10.6% reported ever being pregnant and 4.2% had ever been diagnosed with an STI. Among young men who had ever had sex, 10.5% had ever caused a pregnancy and 3.9% had ever been diagnosed with an STI. The distribution of exposure variables and unadjusted results are outlined in Tables 2 (young women) and 3 (young men); multivariate results are outlined in Table 4.
Among young women who had ever had sex, 32.5% had ever lived on a reserve, 77.0% had learned about culture from their family, and 40.3% reported ever having been sexually abused.
Unadjusted results indicate that having learned about culture from family, ever having lived on a reserve, higher lifetime substance use relative to peers, and ever having been sexually abused were associated with increased odds of pregnancy. Having helped in the community in the past year was associated with decreased odds of pregnancy. For the STI outcome, only higher lifetime substance use was associated with increased odds of STI; feeling more connected to family was associated with decreased odds.
After multivariate modelling, the relationships between pregnancy and sexual abuse, substance use and helping in the community remained significant. For STI, both lifetime substance use and feeling connected to family remained significant.
Among young men who had ever had sex, 37.7% had ever lived on a reserve, 71.3% had learned about culture from their family, and 10.0% reported ever having been sexually abused. Unadjusted estimates show that increased odds of causing a pregnancy were associated with learning about culture from the community, having lived on-reserve, and having been sexually abused. Feeling more connected to school and to family, and having peers who would be angry if you caused a pregnancy were associated with decreased odds of causing a pregnancy. For the STI outcome, using more substances and having been sexually abused were associated with increased odds of STI; feeling more connected to school and having peers who would be angry if you caused a pregnancy were associated with decreased odds of STI.
Multivariate models indicated that living on-reserve, sexual abuse and school connectedness were significant independent predictors of pregnancy involvement. For the STI outcome, lifetime substance use and sexual abuse remained important predictors, and learning about culture from family emerged as a factor associated with increased odds of STI.
This rare glimpse into the sexual health of Aboriginal young people reveals that sexual abuse and substance use are prevalent and strongly associated with self-reported pregnancy and STI diagnosis among Aboriginal young people attending school. Fostering connections to school, family and community may be promising strategies for intervention.
The survey does not represent Aboriginal young people who are not attending school, and our results should not be interpreted as generalizable to this group. Aboriginal young people are more likely than non-Aboriginal young people to leave school after Grade 10, (16) making this particularly important for older adolescents. Self-reported pregnancy has been shown to be more accurate than self-reported STI diagnosis among adolescent girls, (17,18) however, both tend to be underestimates of actual prevalence, particularly for STIs. Obviously, self-reported pregnancy has less validity for young men than for young women; however, no better measure exists. The possibility of type II errors must also be acknowledged, especially in the STI analyses where prevalence of the outcome is low.
Findings in relation to other studies
No previous work has examined relationships between sexual abuse and pregnancy among a probability sample of Canadian Aboriginal adolescents. Our findings of a relationship between sexual health outcomes and sexual abuse (19-22) and substance use (23,24) are consistent with findings from other groups. However, the prevalence of sexual abuse was very high in this sample--some 40% of young women and 10% of young men who have ever had sex report being sexually abused. This is markedly higher than among non-Aboriginal students. (25) Sexual abuse is also highly correlated with potentially risky sexual behaviour among Aboriginal students; (26) thus sexual abuse is extremely important to tackle in risk reduction interventions for Aboriginal students. Similarly, Aboriginal students report higher levels of substance use than non-Aboriginal students. (25) Although the mechanism by which substance use is associated with increased likelihood of these outcomes is subject to debate, (23) it is obviously an important contextual factor which must be accounted for in interventions, regardless of whether or not the association is causal.
At the interpersonal level, community volunteering in the past year was a key predictor of decreased odds of pregnancy for Aboriginal young women but not for Aboriginal young men. In other groups, both positive and negative relationships with community involvement and sexual health have been found, which may be due to the types of organizations people are involved with. (27-29)
School connectedness encompasses feelings of safety and belonging, and is related to positive sexual health outcomes in other populations of young people. (24) For young Aboriginal men, it appears that the school environment also is associated with decreased risk; it did not emerge as a key predictor for young women. In the present sample, however, we would expect a conservative estimate of the effects of school connectedness for young women, since those who were less connected as well as those with children were probably more likely to be absent on the day of the survey.
Instead, for young women, feeling connected to family was related to decreased likelihood of self-reported STI in multivariate analyses. This is consistent with findings from other populations where family connectedness and other family health variables predict decreased sexual risk. (24,30,31) For young men, higher family connectedness decreased the odds of pregnancy in unadjusted analyses but was not a strong independent predictor after other factors were accounted for.
Having lived on-reserve was strongly associated with increased risk of pregnancy involvement among young men, consistent with an increased likelihood of risky sexual behaviours among BC on-reserve young people. (26) Further research is needed to establish why; reasons may include the rural location of most reserves, (32) lower socio-economic status of residents, (33,34) or differing cultural values regarding timing of pregnancy. (4) Learning about culture from the family also emerged as a risk factor in STI analyses, but was only significant in the multivariate model for young men. It is possible that this is simply a statistical artifact owing to low case numbers for the STI analysis, but this link is worth further investigation.
Future directions for programming
Aboriginal young people attending secondary school are clearly in need of sexual risk reduction interventions. At the individual level, students who have been sexually abused and who use more substances are at greater risk of pregnancy and STI; provision of sexual risk reduction programming that addresses history of sexual abuse and substance use may help individuals reduce their risk. However, in an environment where Aboriginal students experience more sexual abuse, (25) use more substances, (25) are generally less connected to school25 and leave school at an earlier age (16) than their non-Aboriginal peers, individualist approaches are likely to yield only limited success. Aboriginal leaders continually call for population-level policy and programming to address the root causes of ill-health; funding and evaluation of these efforts are needed both on- and off-reserve.
Sexual health interventions for in-school Aboriginal young people must address substance use and sexual abuse, and must operate at a broader social level rather than focusing only on treatment/support for individuals.
Received: May 7, 2008
Accepted: January 9, 2009
(1.) Health Canada. A statistical profile on the health of First Nations in Canada. Ottawa, ON: Health Canada, 2002 [updated 2002]. Available online at: www.hc-sc.gc.ca/fnihb-dgspni/fnihb/sppa/hia/publications/statistical_ profile.pdf (Accessed September 29, 2004).
(2.) British Columbia Provincial Health Officer. The Health and Well-being of Aboriginal People in British Columbia. Victoria, BC: BC Ministry of Health Planning, 2002.
(3.) National Center for Health Statistics. Health, United States, 2006 with Chart-book on Trends in the Health of Americans. Hyattsville, MD: NCHS, 2006 [updated 2006]. Available online at: http://www.cdc.gov/nchs/data/hus/ hus06.pdf#004 (Accessed April 8, 2009).
(4.) Devries K. Condom Use and Sexual Health Among Canadian Aboriginal Adolescents. London, UK: University of London, 2007.
(5.) Walters KL, Simoni JM. Reconceptualizing Native women's health: An "indigenist" stress-coping model. Am J Public Health 2002;92(4):520-24.
(6.) Liu LL, Slap GB, Kinsman SB, Khalid N. Pregnancy among American Indian adolescents: Reactions and prenatal care. J Adolesc Health 1994;15(4):336-41.
(7.) Saewyc EM. Influential life contexts and environments for out-of-home pregnant adolescents. J Holistic Nurs 2003;21(4):343-67.
(8.) Saewyc EM, Skay CL, Bearinger LH, Blum RW, Resnick MD. Sexual orientation, sexual behaviors, and pregnancy among American Indian adolescents. J Adolesc Health 1998;23(4):238-47.
(9.) Calzavara LM, Burchell AN, Myers T, Bullock SL, Escobar M, Cockerill R. Condom use among Aboriginal people in Ontario, Canada. Int J STD AIDS 1998;9(5):272-79.
(10.) Spittal PM, Craib KJP, Teegee M. The Cedar project: Prevalence and correlates of HIV infection among young Aboriginal people who use drugs in two Canadian cities. Int J Circumpolar Health 2007;66(3):226-40.
(11.) Wood E, Montaner JS, Li K, Zhang R, Barney L, Strathdee SA, et al. Burden of HIV infection among Aboriginal injection drug users in Canada. Am J Public Health 2008;Epub January 30, 2008.
(12.) Green R. Methodology: Survey Methodology for the 2003 AHS III. Vancouver, BC: McCreary Centre Society, 2003.
(13.) Blum RW, Harris LJ, Resnick MD, Rosenwinkel K. Technical Report on the Adolescent Health Survey. Minneapolis, MN: Adolescent Health Program, University of Minnesota, 1989.
(14.) Sieving RE, Beuhring T, Resnick MD, Bearinger LH, Shew M, Ireland M, et al. Development of adolescent self-report measures from the National Longitudinal Study of Adolescent Health. J Adolesc Health 2001;28(1):73-81.
(15.) StataCorp. Stata Version 9.0. Texas: StataCorp., 2004.
(16.) BC Ministry of Education. Aboriginal Report--How are we doing? Public Schools Only. Victoria, BC: BC Ministry of Education, 2005.
(17.) Clark L, Brasseux C, Richmond D, Getson P, D'Angelo LJ. Are adolescents accurate in self-report of frequencies of sexually transmitted diseases and pregnancies? J Adolesc Health 1997;21:91-96.
(18.) Harrington K, DiClemente RJ, Wingood GM, Crosby RA, Person S, Oh MK, et al. Validity of self-reported sexually transmitted diseases among African-American female adolescents participating in an HIV/STD prevention trial. Sexually Transmitted Dis 2001;28(8):468-71.
(19.) Stevens-Simon C, Reichert S. Sexual abuse, adolescent pregnancy, and child abuse. A developmental approach to an intergenerational cycle. Arch Pediatr Adolesc Med 1994;148(1):23-27.
(20.) Saewyc EM, Magee LL, Pettingell SE. Teenage pregnancy and associated risk behaviors among sexually abused adolescents. Perspectives on Sexual and Reproductive Health 2004;36(3):98-105.
(21.) Anda RF, Chapman DP, Felitti VJ, Edwards V, Williamson DF, Croft JB, et al. Adverse childhood experiences and risk of paternity in teen pregnancy. Obstetrics and Gynecol 2002;100(1):37-45.
(22.) Hillis SD, Anda RF, Felitti VJ, Nordenberg D, Marchbanks PA. Adverse childhood experiences and sexually transmitted diseases in men and women: A retrospective study. Pediatrics 2000;106(1):e11-18.
(23.) Cooper ML. Alcohol use and risky sexual behavior among college students and youth: Evaluating the evidence. J Studies Alcohol 2002;14(Suppl.):101-17.
(24.) Resnick MD, Bearman PS, Blum RW, Bauman KE, Harris KM, Jones J, et al. Protecting adolescents from harm. Findings from the National Longitudinal Study on Adolescent Health. JAMA 1997;278(10):823-32.
(25.) van der Woerd KA, Dixon BL, McDiarmid T, Chittenden M, Murphy A. The McCreary Centre Society. Raven's Children II: Aboriginal Youth Health in BC. Vancouver, BC: The McCreary Centre Society, 2005.
(26.) Devries KM, Free C, Morison L, Saewyc EM. Factors associated with sexual behavior of Aboriginal youth: Implications for health promotion. Am J Public Health 2008;10.2105/AJPH.2007.132597.
(27.) Campbell C, Williams B, Gilgen D. Is social capital a useful conceptual tool for exploring community level influences on HIV infection? An exploratory case study from South Africa. AIDS Care 2002;14(1):41-54.
(28.) Hellerstadt WL, Peterson-Hickey M, Rhodes KL, Garwick A. Environmental, social and personal correlates of having ever had sexual intercourse among American Indian youths. Am J Public Health 2006;96(12):2228-34.
(29.) Crosby RA, DiClemente RJ, Wingood GM, Harrington K, Davies SL, Malow R. Participation by African-American adolescent females in social organizations: Associations with HIV-protective behaviors. Ethnicity & Disease 2002;12(2):186-92.
(30.) DiClemente RJ, Wingood GM, Crosby R, Cobb BK, Harrington K, Davies SL. Parent-adolescent communication and sexual risk behaviors among African American adolescent females. J Pediatrics 2001;139(3):407-12.
(31.) DiClemente RJ, Wingood GM, Crosby R, Sionean C, Cobb BK, Harrington K, et al. Parental monitoring: Association with adolescents' risk behaviors. Pediatrics 2001;107(6):1363-68.
(32.) Crosby RA, Yarber W, DiClemente RJ, Wingood GM, Meyerson B. HIV-associated histories, perceptions, and practices among low-income African-American women: Does rural residence matter? Am J Public Health 2002;92(4):655-59.
(33.) Statistics Canada. 1996 Census: Sources of income, earnings and total income, and family income. Ottawa: Statistics Canada, 1998.
(34.) Santelli JS, Lowry R, Brener ND, Robin L. The association of sexual behaviors with socioeconomic status, family structure, and race/ethnicity among US adolescents. Am J Public Health 2000;90(10):1582-88.
Karen M. Devries, PhD,  Caroline J. Free, PhD, MBChB,  Linda Morison, MSc,  Elizabeth Saewyc, PhD, RN 
[1.] Post-doctoral Fellow, London School of Hygiene and Tropical Medicine, London, UK
[2.] Clinical Lecturer, London School of Hygiene and Tropical Medicine, London, UK
[3.] Senior Lecturer, London School of Hygiene and Tropical Medicine, London, UK
[4.] Associate Professor, University of British Columbia, and Research Director, McCreary Centre Society, Vancouver, BC
Correspondence and reprint requests: Dr. Karen M. Devries, Gender Violence and Health Centre, LSHTM, London International Development Centre, 36 Gordon Square, London WC1H 0PD, Tel: +44 (0) 20 7958 8164, E-mail: email@example.com
Acknowledgements: K. Devries gratefully acknowledges the mentorship of Deborah Schwartz, BC Ministry of Health, and the McCreary Centre Society for access to the BC AHS data. K. Devries was supported by a Canadian Institutes of Health Research Doctoral Research Award and a British Council Overseas Student Research Award at the time of this research.
Table 1. Measurement of Exposure Variables Factor Variable Measurement Age What is your age in years? Continuous Cultural How much have you learned Some or a knowledge-Family about culture from your lot/none family? Cultural How much have you learned Some or a knowledge-School about culture from your lot/none school? Cultural How much have you learned Some or a knowledge-Community about culture from your lot/none community? Reserve Have you ever lived Ever/never on-reserve? Community In the past 12 months, Yes/no involvement-Community did you help others without pay by: helping out in your community? Community In the past 12 months, did Yes/no involvement-Neighbours you help others without pay by: helping neighbours or relatives? School connectedness 7-item scale, e.g., How Continuous much do you feel that your teachers care about you? Peer attitudes Would your peers be angry Yes/no towards pregnancy if you were involved in a pregnancy? Family connectedness 11-item scale, e.g., How Continuous close do you feel to your mother? Sexual abuse 2-item measure: Have you Yes for ever been forced to have selecting sexual intercourse when either/no you did not want to?; Have you ever been sexually abused? Lifetime 11-item measure: Lifetime High/low substance use frequency of using: (Dichotomized Alcohol; Marijuana; on median) Cocaine; Hallucinogens; Mushrooms; Inhalants; Amphetamines; Heroin; Injected an illegal drug; Steroids without a doctor's permission; Prescription pills without a doctor's consent. Table 2. Distributions of Exposures and Unadjusted Associations between Exposure Variables, Ever Being Pregnant and Ever Having an STI among BC Aboriginal Young Women (n=438 *) Full Ever Been Pregnant Sample Exposure Variable % or % or OR 95% CI mean mean ([dagger]) ([dagger]) Age (years) 15.8 16.2 1.22 0.89-1.67 ([double dagger]) Structural factors Have learned No 23.0% 4.8 1 about culture Yes 77.0% 12.4 2.81 1.03-7.61 from family Have learned No 21.4% 13.2 1 about culture Yes 78.6% 10 0.73 0.29-1.80 from school Have learned No 43.8% 7.9 1 about culture Yes 56.2% 12.9 0.74 0.72-4.22 from community Ever lived No 67.5% 7.7 1 on-reserve Yes 32.5% 16.8 2.42 1.05-5.58 Interpersonal factors Helped out in No 64.5% 14.5 1 community, Yes 35.5% 4.5 0.28 0.13-0.62 past year Helped No 35.6% 14.2 1 neighbours, Yes 64.5% 9.3 0.62 0.28-1.40 past year School 34.2% 3.3 0.71 0.43-1.16 connectedness ([double dagger]) (range 1-5) Peers would be No 32.0% 11.6 1 angry if became Yes 68.0% 9.9 0.84 0.40-1.79 pregnant Family 22.9% 2.23 0.73 0.35-1.51 connectedness ([double dagger]) (range 1-3) Individual factors Lifetime Low 47.2% 5.2 1 substance use High 52.8% 15.6 3.38 1.35-8.47 Sexual abuse ever No 59.7% 2.7 1 Yes 40.3% 22.7 10.71 4.29-26.77 Ever Ever Had STI Been Pregnant Exposure Variable p % or OR 95% CI p mean ([dagger]) Age (years) 0.211 16.1 1.19 0.79-1.80 0.410 ([double dagger]) Structural factors Have learned 1.8 1 about culture 0.043 5 2.98 0.87-10.24 0.083 from family Have learned 5.6 1 about culture 0.49 4 0.7 0.22-2.25 0.549 from school Have learned 4.3 1 about culture 0.221 4.4 1.02 0.27-3.94 0.974 from community Ever lived 4.1 1 on-reserve 0.039 4.7 1.15 0.35-3.81 0.821 Interpersonal factors Helped out in 3.9 1 community, 0.002 4.9 1.26 0.28-5.68 0.761 past year Helped 6.5 1 neighbours, 0.249 3 0.44 0.11-1.70 0.233 past year School 0.167 3.41 1 0.48-2.12 0.990 connectedness ([double dagger]) (range 1-5) Peers would be 4.2 1 angry if became 0.655 4.3 1.01 0.33-3.08 0.980 pregnant Family 0.397 1.91 0.21 0.06-0.77 0.018 connectedness ([double dagger]) (range 1-3) Individual factors Lifetime 1.2 1 substance use 0.009 7.1 6.45 1.86-22.43 0.003 Sexual abuse ever 3.4 1 <0.001 5.3 1.58 0.45-5.56 0.478 Analyses based on weighted data and adjusted for survey design. * Actual numbers in each analysis range from 382-438 because of missing data. ([dagger]) Distribution of outcome in each exposure category. Mean for continuous variables (age, school connectedness, family connectedness). ([double dagger]) OR should be interpreted as increase in odds of outcome for every unit increase in exposure score. Table 3. Distributions of Exposures and Unadjusted Associations Between Exposure Variables, Ever Causing a Pregnancy and Ever Having an STI among BC Aboriginal Young Men (n=347 *) Full Ever Caused a Pregnancy Sample Exposure Variable % or % or OR 95% CI mean mean ([dagger]) ([dagger]) Age (years) 15.8 16.0 1.07 0.83-1.39 ([double dagger]) Structural factors Have learned No 28.7% 9.2 1 about culture Yes 71.3% 11.2 1.24 0.56-2.78 from family Have learned No 35.3% 13.9 1 about culture Yes 64.7% 8.8 0.6 0.30-1.18 from school Have learned No 49.0% 7.5 1 about culture Yes 51.0% 13.1 1.86 0.92-3.74 from community Ever lived No 62.3% 5.7 1 on-reserve Yes 37.7% 18.8 3.81 1.78-8.10 Interpersonal factors Helped out in No 75.9% 10.1 1 community, past Yes 24.1% 13.3 1.36 0.59-3.13 year Helped No 42.1% 12.2 1 neighbours, Yes 57.9% 10.0 0.8 0.38-1.66 past year School 345.0% 3.1 0.5 0.32-0.77 connectedness ([double dagger]) (range 1-5) Peers would be No 44.6% 15.4 1 angry if became Yes 55.4% 6.8 0.4 0.19-0.87 pregnant Family 238.0% 2.2 0.46 0.25-0.85 connectedness ([double dagger]) (range 1-3) Individual factors Lifetime Low 46.5% 9.8 1 substance use High 53.5% 10.4 1.07 0.51-2.24 Sexual abuse No 90.0% 8.2 1 ever Yes 10.0% 33.8 2.7 2.46-13.19 Ever Ever Had STI Caused A Pregnancy Exposure Variable p % or OR 95% CI p mean ([dagger]) Age (years) 0.581 15.5 0.88 0.68-1.12 0.292 ([double dagger]) Structural factors Have learned 2.1 1 about culture 0.596 5.1 2.44 0.59-10.08 0.216 from family Have learned 6.6 1 about culture 0.141 2.9 0.43 0.15-1.22 0.114 from school Have learned 3.8 1 about culture 0.084 4.6 1.21 0.40-3.65 0.731 from community Ever lived 3.6 1 on-reserve 0.001 4.9 1.39 0.48-4.03 0.546 Interpersonal factors Helped out in 4.1 1 community, past 0.474 3.3 0.8 0.24-2.70 0.718 year Helped 3.4 1 neighbours, 0.54 4.3 1.29 0.41-3.95 0.653 past year School 0.002 2.93 0.43 0.23-0.80 0.008 connectedness ([double dagger]) (range 1-5) Peers would be 6.1 1 angry if became 0.021 2 0.32 0.09-1.15 0.081 pregnant Family 0.013 2.18 0.49 0.16-1.45 0.196 connectedness ([double dagger]) (range 1-3) Individual factors Lifetime 1.5 1 substance use 0.867 5.4 3.65 1.10-12.17 0.035 Sexual abuse 2.9 1 ever <0.001 14 5.5 1.82-16.61 0.003 Analyses based on weighted data and adjusted for survey design. * Actual numbers in each analysis range from 306-347 because of missing data. ([dagger]) Distribution of outcome in each exposure category. Mean value for continuous variables (age, school connectedness, family connectedness) ([double dagger]) OR should be interpreted as increase in odds of outcome for every unit increase in exposure score. Table 4. Most Important Factors Associated with Pregnancy and STI among Young Aboriginal Men and Women Exposure Variable AOR * 95% CI p Pregnancy Young women Age ([dagger) (years) 1.26 0.94-1.69 0.119 (n=399) Higher lifetime 3.36 1.25-9.08 0.017 substance use Ever been sexually 10.37 4.04-26.60 <0.001 abused Helped out in 0.29 0.12-0.70 0.006 community, past year Young men Age ([dagger]) (years) 1.19 0.90-1.58 0.225 (n=278) School connectedness 0.52 0.32-0.84 0.007 ([dagger]) (range 1-5) Ever been sexually 4.30 1.64-11.25 0.003 abused Ever lived on-reserve 3.60 1.65-7.86 0.001 Ever had STI Young women Age ([dagger]) (years) 1.36 0.92-2.03 0.127 (n=399) Higher lifetime 5.27 1.50-18.42 0.009 substance use Family connectedness 0.22 0.07-0.65 0.006 ([dagger]) (range 1-3) Young men Age ([dagger]) (years) 0.90 0.64-1.26 0.53 (n=297) Higher lifetime 4.60 1.11-19.14 0.035 substance use Learned about culture 7.11 1.47-34.32 0.015 from family Ever been sexually 5.58 1.61-19.37 0.007 abused Analyses based on weighted data and adjusted for survey design. * Adjusted for all other variables in model. ([dagger]) OR should be interpreted as increase in odds of outcome for every unit increase in exposure score.