Reported contraceptive use,risk behaviours and STIs among military conscripts in Estonian defence forces

Abstract

Limited research exists on sexually transmitted infection (STI) and risk behaviour among military personnel. Published research on condom use and types of contraceptives used yield mixed results, yet, the perception that military members are at higher risk for STIs remains. The objectives of this cross-sectional study were to measure factors such as condom use, contraceptive methods, and risky behaviours (i.e. drug use and sex with commercial sex workers) and investigate differences between ethnic groups, where culture could influence behaviour. Data were collected from a recruited population of 584 male, military conscripts in northeastern Europe. Using multinomial logistic regression models, statistically significant findings include an interaction between the use of contraceptive methods of Russians with casual partners and ethnicity, with higher odds of effective methods used among Estonians with regular partners (OR = 8.13) or casual partners (OR = 11.58) and Russians with regular partners (OR = 4.98). Effective contraceptive methods used less frequently with casual partners by ethnic Russians is important in providing education and risk reduction services to young, male conscripts. These findings may be used as a baseline to inform health education and STI prevention programmes tailored to military members in Eastern Europe in the absence of other published studies.

Keywords

Sexual health sexually transmitted infection STI risk factors military sexual behaviour contraception cultural difference Europe

Introduction

Previous research established links between military personnel and higher risk behaviours, including drug misuse.^1–4 The World Health Organization (WHO) reports that military members are more likely to solicit commercial sex workers, though service members are not considered a high risk population.⁵ In the same report, WHO indicates that for sentinel HIV/STI surveillance among commercial sex workers, military members may be used as a proxy. Studies differ on STI prevalence among military personnel.^6–9 Armed forces research identifies factors including age, gender, race, marital status, length of military service, full time or reserve status, and rank modify risk. STI reports from the USA indicate that military membership may not be the mediating factor.⁷ A national voluntary military is non-representative of the general population. Persons from economically disadvantaged backgrounds, lower education and inclined for high-risk behaviours are over-represented and therefore increase risk behaviours and STIs.⁶ Younger service members, women, enlisted personnel, and single marital status persons had higher STI rates across 13 years of surveillance.⁶ Regardless of whether a defence force is volunteer or conscripted, the widely held belief is that military members have higher rates of STIs compared to the non-military population.¹⁰ Even though many militaries conduct compulsory STI testing, estimates are largely unknown due to the limited data.

Reducing risk behaviours among military members requires increasing action in addition to providing education and prevention programmes. Group factors modify behaviour and can change risk related to sexual activities, drug use, contraceptive use, and diet.¹¹ Previous studies consider cultural impacts such as primary language, educational approach, and different behavioural acceptances as important influencers.^12–16 Due to differences between ethnic Estonians and ethnic Russians on STI rates, and the ethnic distribution in Estonia (69.7% Estonian, 25.2% Russian), understanding differences is important to develop effective interventions.^17–19 In 2012, we conducted a prevalence study among male conscripts in Estonia. In addition to screening for the most prevalent, nation-specific STIs, we administered a knowledge, attitude, behaviour, and perception (KABP) assessment to determine levels of risk of sexual and drug-using behaviours.

Methods

This project was reviewed and approved by the Tallinn Medical Research Ethics Committee. We used a cross-sectional study design with convenience sampling to recruited 584 men in July 2012 from four military installations via word of mouth and posted announcements. Participation was completely voluntary. Conscripted service was compulsory for men. Eligibility criteria included: 18 years old, military conscript and the ability to consent. Conscripts received a small chocolate bar, ten condoms and information materials on HIV/STIs as incentives.

Forms were available in both Estonian and Russian. After consent, the self-guided KABP questionnaire was completed. Contraceptive data were in nine categories with researchers designating each as effective or ineffective. The ineffective group consisted of no contraceptive use, calendar method, interrupted intercourse, rinsing, and no knowledge of contraceptive use. The effective group included: contraceptives, spermicide and after-sex pills. Condom use was assessed for anal and vaginal sex in the last 12 months. For STI tests, 10 ml of venous blood and 2 ml of the morning urine were collected. Receiving test results was voluntary and were obtained by presenting a code. Tests included: HIV antibodies + antigen; HCV antibodies (HCV Ab); HBV antigen; Chlamydia – PCR (urine); Gonorrhoea – PCR (urine); Trichomoniasis – PCR (urine). Due to the extremely low national prevalence of syphilis, it was not included for screening. In case of a positive result for HIV, HBV, or HCV, confirmatory tests were offered. Treatment was offered for a positive chlamydia, gonorrhoea, or trichomoniasis result.

Analyses using R or Stata included Fisher’s exact test for categorical variables, Wilcoxon rank-sum test for count variables, test for the equality of proportions and t-test for continuous variables assessed statistical relationships. To capture the possible contraceptive use patterns exhibited by the conscripts, we used a multinomial outcome with three levels: ineffective, effective and mixed contraceptive use (ineffective and effective contraceptive use). Multinomial regression was used to estimate the covariate-outcome associations. An information criterion (AIC) was used for model selection; the AIC uses a function of the likelihood to select the best model from a candidate set of models, given the data that have been observed.²⁰

Results

Conscripts reporting condom use every time in the past 12 months were: 24.0% (n = 106) with regular partners; 43.1% (n = 116) with casual partners; 77.5% (n = 31) with sex workers; 41.2% (n = 7) with injection drug users (IDU); 50.0% (n = 7) with same sex partners; 34.8% (n = 8) for partners with an STI and 41.2% (n = 7) with an HIV-infected partner. Condom use during last sexual intercourse was reported by 45.4% (n = 219) with regular partners; 66.6% (n = 195) with casual partners; 75% (n = 27) with sex workers and 50% (n = 9) with same sex partners. There were few statistically significant differences by ethnicity on condom use with regular partners (P = 0.069). (Table 1). For condom use with a regular partner in the last 12 months, 45.8% (n = 180) of Estonians reported ‘never or seldom’ compared to Russians (66.7%, n = 30). In contrast, casual partner condom use in the previous 12 months reported proportions of ‘never or seldom’ were similar with Estonian (26.3%, n = 63) and Russian (23.1%, n = 6) conscripts, yet no statistically significant associations were identified between ethnicity and the four condom use categories, nor between ethnicity and paid partners, IDUs, partners with an STI or HIV-infected partners.

Table 1.

Demographic characteristics and risk behaviours among conscripted Estonian forces.

	Total (n = 584)^a n (%)	Data by ethnicity^b		P-value^c
	Total (n = 584)^a n (%)	Estonian (n = 523) n (%)	Russian (n = 57) n (%)	P-value^c
Age^d	21.1 ( ±1.37)	21.0 ( ±1.38)	21.7 ( ±1.05)	<0.001
Education^e
Less than secondary	9 (1.6)	5 (1.0)	4 (7.0)	0.010
Secondary	331 (57.1)	295 (56.4)	36 (63.2)
Secondary vocational	180 (31.0)	168 (32.1)	12 (21.1)
Bachelor’s degree	59 (10.2)	54 (10.3)	5 (8.8)
Employed^f	273 (55.3)	242 (54.8)	29 (58.0)	0.602
Unemployed^f	113 (26.7)	98 (25.9)	15 (34.9)	0.232
Student^f	358 (72.2)	320 (72.2)	38 (74.5)	0.506
Have you ever had sex (anal, oral, or vaginal)?^f	532 (91.6)	476 (91.5)	52 (91.2)	0.936
Anal sex^f	106 (18.2)	90 (17.2)	16 (28.1)	0.044
Oral sex^f	522 (89.4)	466 (89.1)	52 (91.2)	0.622
Vaginal sex^f	522 (89.4)	466 (89.1)	52 (91.2)	0. 622
Age at first sex^d	16.97 ( ±1.91)	16.99 ( ±1.90)	16.69 ( ±2.05)	0.285
Condom use at first sexual intercourse^f	385 (76.7)	347 (76.9)	35 (74.5)	0.703
Vaginal or anal sex in the last 12 months^f	481 (90.8)	428 (90.3)	49 (94.2)	0.354
Number of different sexual partners in the last 12 months^g	2 (0–13)	1 (0–13)	2 (1–10)	0.194
Who is supposed to have the condom?
Man	59 (11.1)	47 (9.9)	11 (20.8)	0.008
Woman	8 (1.5)	6 (1.3)	2 (3.8)
Both	463 (87.0)	421 (88.6)	39 (73.6)
Nobody	2 (0.4)	1 (0.2)	1 (1.9)
Did you use alcohol before your last sexual intercourse?
No	222 (41.9)	194 (40.9)	26 (50.0)	0.310
Yes	263 (49.6)	237 (50.0)	24 (46.2)
Do not remember	45 (8.49)	43 (9.1)	2 (3.9)
Did you use illicit drugs before the last sexual intercourse?
No	498 (94.0)	448 (94.5)	46 (88.5)	0.103
Yes	22 (4.2)	17 (3.6)	5 (9.6)
Do not remember	10 (1.9)	9 (1.9)	1 (1.9)
Injection drug use in the last six months^h	3	1	2	–

Numbers do not always add up to 584 due to missing values, percentages due to rounding.

Estonian and Russian counts may not equal the total due to missing ethnicity information.

Fisher’s exact test, Wilcoxon-rank sum test, equality of proportion and t-tests.

Mean ( ± SD).

One Estonian who reported education as a Master’s or PhD.

Frequencies for yes response.

Median (Range).

Percentages and tests not reported due to small numbers.

Multinomial regression explored the relationship between age, ethnicity, education and partner type with contraceptive use. We observed a statistically significant interaction between ethnicity and partner type (Table 2). Table 3 presents this interaction during last vaginal intercourse. Comparing effective to ineffective contraceptive methods, Russian conscripts reported effective methods more with regular than casual partners (OR = 4.98, 95% CI: 1.82–20.20, P = 0.032). Estonians used effective methods more with a regular partner (OR = 8.13, 95% CI: 3.90–22.39, P = 0.003) or casual partner (OR = 11.58, 95% CI: 5.72–30.13, P = 0.001) compared to Russians with a casual partner. When contrasting mixed versus ineffective type, Estonians used mixed methods more than ineffective with a regular partner (OR = 3.76, 95% CI: 1.94–8.25, P = 0.004) or a casual partner (OR = 3.10, 95% CI: 1.60–6.52, P = 0.009) than Russian conscripts with a casual partner. A multinomial logistic model measured associations between age, ethnicity and education on the condom use after STI testing with no statistically significant associations using log odds estimates and AIC model selection.

Table 2.

Contraceptive method used during last vaginal intercourse.

	OR	95% CI	P-value
Effective^a versus ineffective^b
Age	1.09	0.93–1.28	0.285
Ethnicity
Estonian	Reference
Russian	0.62	0.22–1.73	0.364
Education
Secondary	Reference
Secondary vocational	0.75	0.47–1.21	0.238
Bachelor’s degree	0.82	0.40–1.69	0.591
Partner type
Regular	Reference
Casual	1.42	0.88–2.29	0.146
Partner–ethnicity interaction	0.14	0.04–0.55	0.005
Mixed^c use versus ineffective^b
Age	1.07	0.91–1.27	0.419
Ethnicity
Estonian	Reference
Russian	0.28	0.08–0.95	0.041
Education
Secondary	Reference
Secondary vocational	0.96	0.59–1.56	0.869
Bachelor’s degree	0.57	0.25–1.28	0.176
Partner type
Regular	Reference
Casual	0.82	0.50–1.34	0.434
Partner–ethnicity interaction	1.16	0.28–4.85	0.835

Effective contraceptive group: contraceptives, spermicide, and after-sex pills.

Ineffective contraceptive group: no contraceptives, calendar method, interrupted intercourse, rinsing, and unknown partner use.

Mixed contraceptive group reported using both.

Table 3.

Ethnicity-partner type associations^a,b with contraception method for last vaginal intercourse.

	Ethnicity
	Estonian	Russian
Effective^c versus ineffective^d
Regular partner	OR: 8.13 95% CI: 3.90–22.39 P = 0.003	OR: 4.98 95% CI: 1.82–20.20 P = 0.032
Casual partner	OR: 11.58 95% CI: 5.72–30.13 P = 0.001	Reference
Mixed^e use versus ineffective^d
Regular partner	OR: 3.76 95% CI: 1.94–8.25 P = 0.004	OR: 1.05 95% CI: 0.27–3.75 P = 0.498
Casual partner	OR: 3.10 95% CI: 1.60–6.52 P = 0.009	Reference

95% confidence bootstrap intervals derived using the 5% and 95% percentiles.

P-value derived using t-test constructed with the coefficient in the numerator and the bootstrap estimate of the standard error in the denominator. The effective degrees of freedom from the multinomial logistic regression (effective df = 14).

Effective contraceptive group: contraceptives, spermicide, and after-sex pills.

Ineffective contraceptive group: none, calendar method, interrupted intercourse, rinsing, and unknown partner use.

Mixed contraceptive group reported both methods.

There were 29 positive STI results among the conscripts. The mean age of conscripts with an STI diagnosis was 21.4 years (median=22 years, range: 19–23 years). There were no positive results for HIV or HBV, and only a single positive result for HCV. The HCV positive conscript was referred for confirmatory testing in the civilian healthcare system. There were 18 cases of chlamydia (3%), with 17 of the cases reported among the Estonian conscripts; five positive gonorrhoea tests and five positive trichomoniasis tests. Estonian conscripts made up all 10 cases of gonorrhoea and trichomoniasis. None of these conscripts had multiple infections. All conscripts with an STI diagnosis reported never having intercourse with a person of the same sex, having a heterosexual orientation, and having oral sex and vaginal intercourse. Only six of these conscripts reported having had anal sex (21.4%). Only one conscript with an STI reported using illegal drugs before the last time of having sexual intercourse.

Discussion

Condom use among conscripts was similar between ethnic groups, whereas risk behaviour differences were identified. Estonian conscripts were less likely to engage in anal sex, more likely to indicate men and women were responsible for a condom at sexual intercourse, and less likely to inject drugs in the previous six months. Contrary to current literature, condom use after STI testing did not significantly change.^17,21 Despite the extensive educational campaigns and overrepresentation of HIV among ethnic Russians, less reported condom use and low effective contraceptive methods were reported.¹⁸ These conscripts either perceived their risk lower with a regular partner or there is a difference in the actualisation of effective STI prevention. Similar to recent research in other military populations, a higher proportion of conscripts (77.5%) chose to always use a condom when engaging in sexual intercourse with a commercial sex worker.¹⁹

The interaction between ethnicity and contraceptive use across partner type (regular, casual) allowed for a group comparison that could have been lost if the outcome had been reduced to a binary outcome with possible misclassification of the mixed contraceptive use. While not necessarily linked to acquisition of STIs, contraceptive use was previously linked with increased risky sexual behaviours.²² All conscripts utilised effective methods of contraceptives more with a regular partner than with a casual partner. Since condom use was included in the effective contraceptive group, yet was not indicated independently, conscripts might be more concerned with pregnancy of a regular partner. Research indicates that sexual behaviours with casual partners include more high-risk behaviours.²³ This offers an indication of areas for increased education and interventions to promote effective contraceptive methods as well as to explore the risk behaviours surrounding these casual sexual encounters.

These findings are important as research on condom and contraceptive use is limited among militaries, especially among countries in the region with similar epidemics. For nations unable to replicate or conduct such a study, these findings may be useful in their approach to STI education and prevention among conscripts. Lastly, these findings serve as a baseline for the Estonian STI military education and prevention programme. Our methodology provides a snapshot of conscripts in Estonia, but is limited. We were not able to use random sampling due to the need for confidentiality of the conscripts and the strict military schedule. Some small numbers in the analyses could make some assumptions unstable; however, using non-parametric analyses increases the generalisation to all Estonian conscripts. Sexual behaviour and drug use questions may have been prone to recall and social desirability bias. As the conscripts were within 30 days of service, most reported behaviour would have been pre-induction. Use of a more in-depth tool, such as the Risk Behaviour Assessment (RBA), has been found more accurate among military members for condom use.²¹ A larger sample with additional metrics and a prospective design with random sampling could further elucidate potential identifiers for these differences.

Future research includes a similar study on professional military forces. This exploration may yield different findings among condom usage and risk behaviours with this separate voluntary military population. Given the findings of this study, it is recommended to continue military education on safer sexual practices and discuss how these practices can decrease the risk of STIs.

Footnotes

Acknowledgements

The authors thank the Estonian Defense Forces and Estonian Ministry of Defense for their support and assistance.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by the US Naval Health Sciences Research Center through a grant under the Defense HIV/AIDS Prevention Program.

References

National Research Council. Substance use disorders in the U.S. armed forces—Institute of Medicine, Washington, DC: National Research Council, 2012.

Harbertson

Grillo

Zimulinda

. HIV seroprevalence, associated risk behavior, and alcohol use among male Rwanda defense forces military personnel. AIDS Behav 2012, pp. 1–12.

Chun

Tavarez

Dann

. Interviewer gender and self-reported sexual behavior and mental health among male military personnel. Int J Public Health 2011; 56: 225–229.

Anastario

Manzanero

Blanco

. HIV infection, sexual risk behaviour and condom use in the Belize defense force. Int J STD AIDS 2011; 22: 73–79.

Guidelines on surveillance among populations most at risk for HIV. Geneva: UNAIDS/World Health Organization, 2011.

Armed Forces Health Surveillance C. Sexually transmitted infections, active component, U.S. Armed Forces, 2000–2012. Msmr 2013; 20: 5–10.

Spaulding

Lifson Alan

. Gonorrhoea or chlamydia in a US military HIV-positive cohort. Sex Transm Infect 2012; 8: 266–271.

Brown

Ross

Simpson

. Prevalence of markers for HIV, hepatitis B and hepatitis C infection in UK military recruits. Epidemiol Infect 2011; 139: 1166–1171.

Singer

Bautista

O'Connell

. HIV infection among US Army and Air Force military personnel: sociodemographic and genotyping analysis. AIDS Res Human Retrovirus 2010; 26: 889–894.

10.

Feldbaum

Lee

Patel

. The national security implications of HIV/AIDS. PLoS Med 2006; 3: e171–e171.

11.

Brar

Shuter

Thomas

. A comparison of factors associated with prevalent diabetes mellitus among HIV-Infected antiretroviral-naive individuals versus individuals in the National Health and Nutritional Examination Survey cohort. J Acquir Immune Defic Syndr 2007; 45: 66–71.

12.

Marshall

Kerr

Shoveller

. Structural factors associated with an increased risk of HIV and sexually transmitted infection transmission among street-involved youth. BMC Public Health 2009; 9: 7–7.

13.

Rhodes

Hergenrather

Aronson

. Latino men who have sex with men and HIV in the rural south-eastern USA: findings from ethnographic in-depth interviews. Cult Health Sex 2010; 12: 797–812.

14.

Rhodes

Kelley

Siman

. Using community-based participatory research (CBPR) to develop a community-level HIV prevention intervention for Latinas: a local response to a global challenge. Women's Health Issue: Official Publication Jacobs Inst Women's Health 2012; 22: e293–e301.

15.

Sapiano

Moore

Kalayil

. Evaluation of an HIV prevention intervention designed for African American Women: results from the SISTA community-based organization behavioral outcomes project. AIDS Behav 2013; 17: 1052–1067.

16.

Merrill

Hedin

Fondario

. Drug poisoning deaths according to ethnicity in Utah. J Addiction 2013; 2013: 380161–380161.

17.

Bachmann

Grimley

Gao

. Impact of a computer-assisted, provider-delivered intervention on sexual risk behaviors in HIV-positive men who have sex with men (MSM) in a primary care setting. AIDS Educ Prev 2013; 25: 87–101.

18.

Uuskula

Kals

Rajaleid

. High-prevalence and high-estimated incidence of HIV infection among new injecting drug users in Estonia: need for large scale prevention programs. J Public Health (Oxf) 2008; 30: 119–125.

19.

Hallum-Montes

D'Souza

Tavarez

. Condom use during last sexual contact and last 30 days in two samples of Caribbean military personnel. Am J Men Health 2012; 6: 132–135.

20.

Akaike H. Information theory and an extension of the maximum likelihood principle. In: Parzen E, Tanabe K, Kitagawa G, editors. Selected Papers of Hirotugu Akaike. New York: Springer Series in Statistics: Springer, 1998: 199–213.

21.

Tobin

Kuramoto

German

. Unity in diversity: results of a randomized clinical culturally tailored pilot HIV prevention intervention trial in Baltimore, Maryland, for African American men who have sex with men. Health Educ Behav 2013; 40: 286–295.

22.

Wilson

Uuskula

Feldman

. A case-control study of beliefs and behaviors associated with sexually transmitted disease occurrence in Estonia. Sex Transm Dis 2001; 28: 624–629.

23.

Abdala

Zhan

Shaboltas

. Efficacy of a brief HIV prevention counseling intervention among STI clinic patients in Russia: a randomized controlled trial. AIDS Behav 2013; 17: 1016–1024.