CYBERPSYCHOLOGY, BEHAVIOR, AND SOCIAL NETWORKING Volume 18, Number 2, 2015 ª Mary Ann Liebert, Inc. DOI: 10.1089/cyber.2014.0343

ORIGINAL ARTICLES

Association Between Pornography Use and Sexual Risk Behaviors in Adult Consumers: A Systematic Review Emily L. Harkness, B Psych (hons),1 Barbara M. Mullan, PhD,1,2 and Alex Blaszczynski, PhD1

Abstract

The purpose of this review was to determine whether an association exists between sexual risk behaviors and pornography consumption. Consumption of pornography is common, yet research examining its link with sexual risk behaviors is in its infancy. Indicators of sexual risk behavior, including unsafe sex practices and a higher number of sexual partners, have been linked to poor health outcomes. A systematic literature search was performed using Medline, PsycINFO, Web of Knowledge, Pubmed, and CINAHL. Studies were included if they assessed the association between pornography use and indicators of sexual risk behaviors in an adult population. A total of 17 were included in the review, and all were assessed for research standards using the Quality Index Scale. For both Internet pornography and general pornography, links with greater unsafe sex practices and number of sexual partners were identified. Limitations of the literature, including low external validity and poor study design, restrict the generalizability of the findings. Accordingly, replication and more rigorous methods are recommended for future research. texts, and is widely available both on- and offline.4 Internet pornography is believed to represent a unique form of sexually explicit media, which is qualitatively different from non-Internet pornography.5 Cooper6 proposed the Triple A Engine theory, which suggests that the affordability, accessibility, and anonymity of the Internet creates a powerful medium for expressing one’s sexuality and interacting with sexual material. While there has been mixed evidence and limited support for the Triple A Engine theory, it serves to emphasize that findings in relation to sexual behavior may differ between offline and Internet pornography,7,8 which warrants investigation. Some correlates of both online and offline pornography use have been widely researched. However, research has predominantly focused on sexist attitudes, sexual violence, addiction, victimization through illegal pornography, threats to family values, acceptance of rape myths, and negative role models.5,9 Association between pornography use and risky sexual behavior has received limited attention, despite its relevance as an important sexual health issue.5 While there is a range of explanations addressing how pornography consumption may influence sexual behavior, sexual script theory10 is the dominant perspective. Sexual scripts are cognitive schemata representing internalized

Introduction

T

he production and consumption of pornography, also referred to as ‘‘sexually explicit material,’’ is widespread. In 2006, $3.62 billion was spent on adult video sales and rentals in the United States alone, with overall pornography revenue estimated at $13.33 billion.1 In Australia, pornography revenues were estimated at $2 billion.1 However, as this corresponded to a relatively smaller population, per adult spending is estimated to be considerably higher. No standardized definitions of pornography exist, and those available are variable and inconsistent.2 Recently, a systematic review of Internet pornography research methodology concluded that definitions should include the type of, and motivation for, pornography viewed, broadly defining pornography as ‘‘any sexually explicit material displaying genitalia with the aim of sexual arousal or fantasy.’’2(p21) Nonetheless, the most commonly employed definition seems to be that of Hald and Malamuth who defined pornography as ‘‘any kind of material aiming at creating or enhancing sexual feelings or thoughts in the recipient and, at the same time, containing explicit exposure and/or descriptions of the genitals and clear explicit sexual acts.’’3(p616) Pornography may come in a variety of media, including magazines, books, video clips, films, photos, comics and 1 2

School of Psychology, University of Sydney, Sydney, Australia. School of Psychology and Speech Pathology, Curtin University, Perth, Australia.

59

60

social messages that guide sexual decision making, desires, expectations, and behavior in sexual interactions.10,11 As pornography may inform sexual socialization, it could be an important source through which sexual scripts regarding condom use and norms around casual sex are acquired, modified, and acted upon.12,13 Another popular view is that of social cognitive theory. Bandura,14 suggested that individuals may learn unsafe sex behaviors through modeling. Specifically, pornographic depiction of explicit rewards associated with unsafe sex behaviors (e.g., male orgasm and gratification), coupled with an absence of associated punishments or negative consequences,15 may increase the desirability and likelihood of enacting this behavior. Research into the effects of pornography use on sexual risk behavior has largely been conducted in adolescents, with evidence for associations between exposure to pornographic Web sites and a higher number of lifetime sexual partners, having multiple sexual partners in the previous 3–6 months, not using a condom during the last sexual encounter, and not using contraception in the previous 6 months (e.g., BraunCourville and Rojas, Luder et al., and Wingood et al.16–18). Adolescence is a critical period for psychosexual development, as sexual socialization may have a strong impact on the formation of sexual scripts during this time. In contrast, it is often assumed that adults are less susceptible to the influences of pornography, due to their enhanced critical thinking skills and well-established sexual scripts. There has been substantially less research undertaken for adult pornography consumers to investigate this idea,19 despite its potential importance and clinical significance. Research has indicated that pornography may be a significant source of sexual education, particularly where other information sources are unavailable.20 Nevertheless, safesex messages and practices are inconsistent and often absent in pornography. Furthermore, as a result of rising online amateur pornographic productions, there is reduced control over the depiction of safe-sex practices.21 There has been little direct research into how such messages may affect adult consumers’ safer-sex attitudes and practice.21 This is a crucial area of investigation, as safer sex behavior has a direct impact upon risk of contracting sexually transmitted infections (STIs). A review of the research into the association between safer-sex practices and pornography use will assist in highlighting this area as an important current issue within the adult population, and improve the cohesion of the literature, informing directions for future research. Other indicators of sexual risk behavior, such as a higher number of sexual partners and casual sex behaviors, may also be associated with pornography use. Increased risk of STI transmission has been linked to casual sex, defined as sex between uncommitted persons such as acquaintances, first dates, or sex workers and clients,22 and a higher number of sexual partners.16,23 Rationale and aims of the current systematic review

Sexual risk behaviors may have serious consequences for physical, psychological, and sexual health. Understanding the nature of the relationship between sexual risk taking and pornography consumption is therefore crucial to illuminate further the psychological determinants of STIs.23 Furthermore, knowledge of this relationship, if it exists, may assist

HARKNESS ET AL.

effective development and implementation of sexual health interventions, such as STI prevention.24 Accordingly, the aim of the current systematic review was to summarize the research examining the association between pornography use and indicators of sexual risk behaviors, including: (a) unsafe sexual practices and condom use, (b) number of sexual partners, (c) casual sex practices, and (d) STI incidence. Considering the psychosexual differences between adults and adolescents, the focus of the current review was restricted to adult consumers. Due to the unique nature of the Internet, an additional aim was to determine whether the association between indicators of sexual risk behaviors and Internet pornography use was significantly different from that of unspecifieda pornography use. Methods

A literature search of published studies was conducted in March 2014. The electronic databases Medline, PsycINFO, Web of Science, Pubmed, and CINAHL were searched. Keywords associated with pornography (sexually explicit material, sexually explicit media, pornography, porn*, cyberporn, online pornography, Internet pornography, online erotica, erotica, Internet erotica, and cyberpornography) were combined with terms associated with indicators of sexual risk behaviors (sexual risk behavior, risk behavior, sexual risk taking, risk taking, condom/condoms, safe sex, unsafe sex, safer sex, unprotected sex, contraception, birth control, casual sex, sexual partners, HIV, AIDS, STIs, or sexually transmitted diseases [STDs]). A manual search was also performed, examining the references of relevant articles for any studies that may have been overlooked by the database search if obscure terminology was used. To meet inclusion criteria, studies needed to: (a) include adults older than 18 years of age; (b) examine legal pornography use (amount of use, age at first exposure, problematic use); (c) examine indicators of sexual risk behaviors in relation to pornography use (unsafe sex, casual sex, number of sexual partners, STI incidence); (d) be a randomized controlled trial (RCT), cohort, correlational, quasi-experimental, cross-sectional, or qualitative study; (e) be available in English; and (f) have the full text available. Articles were excluded if the sample included children or adolescents. Only works published in peer-reviewed journals or books were included; dissertations were deemed eligible for inclusion but none were identified through the search strategy. Clinical case studies, commentaries, theoretical essays, or narratives and other nonempirical articles were excluded. While legal paraphilic pornography use, such as that of sadomasochism or bondage, was included, nonconsensual violent or illegal sexual acts (such as rape, pedophilia, and bestiality) were excluded. Further, attitudes toward risky behaviors were not the focus of this review, and were excluded. Additionally, problematic Internet use or pornography addiction were not included as outcome variables, as the focus of the review was restricted to physical sexual health risk behaviors. Figure 1 displays the study selection process for the review. In total, 774 articles (excluding duplicates) were identified through the search strategy. All titles and abstracts were initially screened to determine relevance. A second researcher screened a random selection of 10% of the titles and abstracts in order to ensure interrater reliability, with

PORNOGRAPHY USE AND SEXUAL RISK BEHAVIORS

61

FIG. 1. Flow chart for article selection.

100% agreement obtained. The full text articles were then further assessed for eligibility; the reasons for exclusion at this stage are detailed in Figure 1. In total, 17 studies were retained for final review; it is of note that no included articles examined pornography use in relation to STI incidence. Articles were assessed for quality utilizing a modified version of the Quality Index Scale (QIS),25 adapted to allow for the appraisal of nonintervention studies.26 The QIS has good reliability and validity for measuring the methodological quality of health research.25,27–29 Individual items were scored either 0 (no/unable to determine) or 1 (yes). The standard of reporting (possible range 0–7), external validity (possible range 0–3), and internal validity (possible range 0– 4) and study power (possible range 0–1) were assessed. The maximum score obtainable on this scale is 15, with higher scores indicating greater methodological quality. Results

Of the 17 studies that met the criteria for inclusion, two articles examined Internet pornography and a further 15 articles did not differentiate the type of pornography used. For

comparative purposes, the findings for Internet pornography and unspecified pornography (herein referred to simply as ‘‘pornography’’) will be considered separately. Details of the articles reviewed are summarized in Table 1 (Internet pornography studies) and Table 2 (pornography studies). The individual means and standard deviations (SD) for the QIS are displayed in Table 3 (Internet pornography studies) and Table 4 (pornography studies). The total mean QIS was 9.82 (range 6–13) across both pornography and Internet pornography categories. Unsafe sex practices and condom use

Seven studies,23,30–35 including the two Internet pornography studies, investigated the association between pornography use and safer sex practices. All seven studies found evidence of an association. However, findings were mixed. In a community sample of HIV-negative men who have sex with men (MSM), Eaton et al.23 found that time spent viewing pornography was associated with the number of unprotected insertive anal acts, but not the number of unprotected receptive anal sex acts in the previous month. In a similar study, Stein

62

Type of study Gender

Cohort

Sexual orientation Relevant findings

Effect size

QIS

MSM 18–29: 45%, 30–39: 13%: 40–49: 19%, 50 + : 23% (M, SD, and range NR)

Percentage of IP viewed showing UAI

UAI in the past Percentage of IP depicting UAI 3 months, significantly serodiscordant and positively UAI in related to the past engaging in 3 months UAI and serodiscordant UAI in the previous 3 months

a When controlling for gender, age, sensation seeking, life satisfaction, attachment, heterosexual orientation, relationship status, number of lifetime sexual partners, and sexual risk behaviors of friends. *p < 0.05; **p < 0.01; ***p < 0.001. QIS, Quality Index Scale; IP, Internet pornography; UAI, unprotected anal intercourse; NR, not reported.

Peter and Valkenburg (2011); Netherlands

Mean age (SD; range)

7 UAI: 0–24% of IP showing UAI: OR = 1, 25–49%: OR = 1.7, 50–74%: OR = 2.9, 75–100%: OR = 4.9; Serodiscordant UAI: 0–24% of IP showing UAI: OR = 1, 25–49%: OR = 1.4, 50–74%: OR = 2.0, 75–100%: OR = 2.7 Random Frequency of OR = 2.201; Frequency 51% female, 47.89 years 91% Viewing 13 community viewing IP r = 0.09**–0.14** of casual 49% male (16.67; range ‘‘exclusively frequency sample correlated with (across two time sex without unavailable) heterosexual’’ of IP over a higher points); for gender using a the past (n = 833 number of interaction: condom in 6 months adults) unsafe casual OR = 1.43 · 1024 past 6 months sex acts** over the previous 6 months.a A gender–IP use interaction was also significant, and post hoc analyses revealed that the effect was only significant for only males, and IP use not associated with female casual unsafe sex.

Population (n)

Male Nelson CrossMSM who et al. (2014); sectional accessed United States an MSMseeking Web site (n = 1,170)

Author, year, and country

Relevant outcome Pornography variable(s) and variable measure(s)

Table 1. Summary of Findings for Internet Pornography

63

Kraus and Russell (2008); United States Morgan (2011); United States

CrossCollege students sectional (n = 782)

58% female, 19.9 (SD NR; 42% male 18–30)

CrossCommunity sample, Males 29.1 years (10; sectional HIV negative, range NR) had q2 unprotected male anal sex partners in past 6 months (n = 149) CrossConvenience 63% female, 30.0 (11.8; sectional community sample 37% male range NR) (n = 437)

62% female, 20.0 years 38% male (1.84; 18–26)

Gender

Eaton et al. (2012); United States

Population (n)

Mean age (SD; range)

CrossUniversity students sectional (n = 813)

Type of study

Carroll et al. (2008); United States

Author, year, and country

Heterosexual

92% Heterosexual

Homosexual

Lifetime # sexual partners, past 12 months # sexual partners

Relevant outcome variable(s) and measure(s) Relevant findings

Frequency of pornography use associated with a higher # lifetime sexual partners (males,* females***) and # sexual partners over past 12 months (females***only, males ns); pornography users had more sexual partners across lifetime* and past 12 months* than nonusers. Daily male users had a higher # of lifetime sexual partners* than none or seldom users, but lower # of sexual partners over past 12 months than nonusers.*a Monthly users had a greater # of partnersa from past 12 months than nonusers*. Female users had higher # of partners over past 12 months* and lifetime* compared to nonusers.a Average minutes # Male sexual Pornography viewing time per per week partners, # week significantly associated viewing receptive UAI, # with a higher # of male sexual pornography insertive UAI partners*** and insertive UAI*** but not with # of receptive UAIb History of Lifetime number of No differences found between those exposure to sexual partners exposed to SEM between age 12 pornography for and 17 and those not for lifetime ages 12–17 # sexual partners ( p > 0.05) Frequency of Lifetime # sexual Frequency of pornography used pornography partners, # was associated with # of lifetime use, type of lifetime casual casual sex partners** and with # pornography sex partners of lifetime sexual partners**,c Type of pornography used was also viewed associated with # of lifetime casual sex partners** and with # of lifetime sexual partners**,c Frequency of pornography use, as well as type of pornography used, were significantly associated with a higher # of lifetime sexual partners in males** and females,*** and with a higher # of lifetime casual sex partners in both males*** and females***,c

Pornography variable

96% heterosexual, Pornography 2% homosexual, viewing 2% bisexual frequency

Sexual orientation

Table 2. Summary of Findings for Unspecified Pornography

8

Lifetime # partners: frequency of use (b = 0.15), males r = 0.20, females r = 0.38, type used (b = 0.28), males r = 0.29, females r = 0.39; # lifetime casual partners: frequency (b = 0.12), males r = 0.20, females r = 0.32; type used (b = 0.30), males r = 0.30 females r = 0.36

(continued)

6

13

NR

NR

9

QIS

Lifetime # partners: males r = 0.11, females r = 0.17; Past 12 months # partners: males r = 0.06, females r = 0.09; other data NR

Effect size

64 Males

CrossCommunity sample of sectional nonmonogamous MSM (n = 821)

Stein et al. (2012); United States

Rosser et al. (2013); United States

CrossProbabilistic, multistage 49.3% sectional stratified national female, sample (n = 1,005) 50.7% male

Gender

Sinkovic et al. (2013); Croatia

Population (n)

CrossCommunity-based 50% female sectional sample of 617 couples (n = 1,234); referred by therapists, relationship educators, clergy, family or friends, or online adverts CrossOnline community Males sectional sample (n = 1,391)

Type of study

Poulsen et al. (2012); United States

Author, year, and country

Mean NR, median = 32 (18–68)

NR (NR; 18–25)

NR; all participants were 18 + , mode = 18–24

NR (NR; 17–58)

Mean age (SD; range)

90.7% gay identified

94.2% Heterosexual

MSM, gay identified

Heterosexual

Sexual orientation

Relevant outcome variable(s) and measure(s) For both males and females, pornography users had a significantly greater number of sexual partners than nonusers

Relevant findings

Viewing hours per # UAI male partners Receptive UAId: marginal correlation. Linear trend week, frequency for both receptive significant*: rate increased as of viewing and insertive acts. viewing time increased. Insertive pornography Whether UAIc: correlation/linear trend not depicting participant had significant. Serodiscordant UAId: unprotected vs. engaged in correlation not significant. protected anal serodiscordant or Positive linear trend significant.* intercourse potentially Preference for UAI porn increased serodiscordant risk behavior* and preference for UAI (yes/no) viewing safe anal intercourse had significantly less risk behavior, compared to no preference.* Age at first Sexual risk taking Sexual risk taking index: females— exposure to (index of: early negatively correlated with age at pornography, sexual debut < 15 first exposure** and not frequency of years), condom significantly correlated with viewing over use during last frequency of use; males— previous 12 intercourse, negatively correlated with age at months consistent first exposure*** and not condom use over significantly correlated with past month, frequency of use; age at first concurrent sexual exposure*** but not frequency of relationships, use significant predictor in intercourse with regression analysise unknown person) Proportion of UAI behavior in Viewing pornography that depicted pornography prior 3 months UAI was significantly associated viewed that (receptive anal with engaging in UAI***; depicted UAI sex acts, insertive those who viewed UAI pornography anal sex acts, or 25–74%** and 75–100%*** of both) the time had an increased UAI (insertive) acts in the previous 3 months when compared to those who only viewed UAI pornography 0–24% of the time. The findings were similar for UAI receptive acts (25–74%* and 75– 100%*** when compared to those viewing UAI pornography 0–24% of the time) and for both insertive and receptive UAI acts (25– 74%*** and 75–100%*** when compared to those viewing UAI pornography 0–24% of the time)

Frequency of # lifetime sexual pornography partners use over the past 12 months (ranging from never to almost every day)

Pornography variable

Table 2. (Continued)

12

Sexual risk taking index: age at first exposure b = 0.14

(continued)

UAI (insertive): 12 UAI pornography 25–74% OR = 1.9; 75–100% OR = 4.4 UAI (receptive): UAI pornography 25–74% OR = 1.7; 75–100% OR = 3.5 UAI (both): UAI pornography 25– 74% OR = 2.9, 75–100% OR = 8.1

9

7

QIS

NR

Males: Cohen’s d = 0.44; females: d = 0.30

Effect size

65

Weinberg et al. (2010); United States

Træen and Daneback (2013); Norway

Sˇtulhofer et al. (2010); Croatia

Author, year, and country

Population (n)

Gender

Mean age (SD; range)

CrossCollege students in a sectional sociology course (n = 172) 59% female, 41% male

CrossSexually active college Males NR, range 18–25 sectional students/youth who have used pornography (n = 650) CrossRandom sample drawn 59% female, NR (NR; 18–59) sectional from population 41% male register (n = 2,381)

Type of study

Pornography variable

Relevant outcome variable(s) and measure(s) Relevant findings

Effect size

QIS

(continued)

Paraphilic Lifetime # sexual Paraphilic pornography users had NR 10 pornography partners an increased # of lifetime sexual use (last 12 partners* than users of months) nonparaphilic pornography 95% heterosexual, % of time 7 Number of lifetime, % time viewing pornography while % time viewing 5% gay/lesbian/ watching past 5 years, 3 masturbating: Heterosexual pornography bisexual pornography years, 12 months women: lifetime male,*** 5 while while of male and years male,*** 3 years male,*** masturbating: masturbating female partners 12 months male*** (n.s. for all Heterosexual and % time others); Lesbian/bisexual women: lifetime viewing women: n.s. for all; Heterosexual male r = 0.16, 5 years male pornography men: 5 years female,** 3 years r = 0.17, 3 years during sex (past female,** 12 months female,* male r = 0.17, 12 months) 3 years male* (n.s. for all others), 12 months male gay/bisexual men: n.s. for all r = 0.16; % time viewing pornography during sex: Heterosexual women: n.s. Heterosexual men: 5 years female for all; lesbian/bisexual women: r = 0.12, 3 years 3 years female* (n.s. for all female r = 0.12, others); heterosexual men: 12 months female 5 years female,* 5 years male,* r = 0.09, 3 years 3 years male,** 12 months male r = 0.36 male*** (n.s. for all others), % time viewing gay/bisexual men: n.s. for all; pornography during sex: lesbian/bisexual women: 3 years female r = 0.45; heterosexual men: 5 years female r = 0.10, 5 years male r = 0.43, 3 years male r = 0.50, 12 months male r = 0.64 heterosexual 6 Females: 68% Frequency of # sexual partners Pornography use associated with females: for heterosexual, pornography over past 12 increased # sexual partners in (e): b = 0.39, 32% viewing over months where past 12 months for those they OR = 1.48; Nonnonheterosexual; the past 12 they (a) didn’t were currently significantly heterosexual males: 73% months (X-rated know them/just involved with** for heterosexual females: for heterosexual, movies, met, (b) friend or females (n.s. for all other types of (d): b = 0.92, 27% pornography in acquaintance, (c) partner). Pornography use associated with increased # OR = 2.50 for nonheterosexual magazines, or were on a first sexual partners in past 12 months (e): b = 0.54, IP) date with, (d) had for those they had previously OR = 1.72 previously dated, dated* and those they were or (e) were currently significantly involved significantly with* (n.s. for all other types of involved with partner). However, no associations for heterosexual and nonheterosexual males (all n.s.).

NR

Sexual orientation

Table 2. (Continued)

66

CrossGeneral social survey sectional data from 23 time points (independent samples) during 1973–2010 (n = 18, 225) Females

Males

58%

Gender

NR

30.1 (NR; 18–49)

NR

46.1 (17.9; 18–89) NR

44.73 (17.03; 18–89)

T1: 45.0 (16.8, NR 18–89) T2: 47.2 (16.8, 19–89)

Mean age (SD; range)

Sexual orientation

Relevant outcome variable(s) and measure(s) Relevant findings

Effect size

Pornography Engagement in exposure in the casual sex in the past 12 months past year (yes/no)

Pornography exposure at T1 was OR = 1.92 associated with increase in odds of casual sex at T2, after controlling for T1 casual sex behavior # partners 12 Pornography # sexual partners Pornography use positively months: b = 0.17, exposure in the over past 12 associated with # sexual partners e r = 0.25 (1980s past 12 months months (14 data in past 12 months *** and past e 5 years ***; unmarried r = 0.29, 1990s (yes/no) points), # over pornography users were more r = 0.20, 2000s past 5 years (11 likely to use a condom than r = 0.27) data points), nonusers** # partners 5 years: condom use b = 0.20, r = 0.32 during last (1990s r = 0.26, encounter (9 data 2000s r = 0.34); points) condom use: r = –0.06 # partners 12 Pornography # sexual partners Pornography use positively months: b = 0.12, exposure in the over past 12 associated with # sexual partners r = 0.24; (1980s past 12 months months (14 data in past 12 monthse** and past e 5 years ** r = 0.21; 1990s: (yes/no) points), # sexual r = 0.22; 2000s partners over past r = 0.26); # 5 years (11 data partners 5 years: points) b = 0.16, r = 0.28; (1990s r = 0.24; 2000s r = 0.32) Frequency of Having had multiple Odds of having multiple sexual Frequently use pornography sexual partners partners higher for greater pornography: use frequency of pornography use** OR = 1.00, sometimes use: OR = 0.67, never use: OR = 0.31

Pornography variable

11

13

13

11

QIS

c

b

Controlling for age, dating status, religiosity, and impulsivity. Controlling for age, income, substance use (sex drugs and alcohol), perceived risk of unprotected anal intercourse, and condom use self-efficacy. Controlling for gender, religiosity, and dating status. d Controlling for age, education, ethnicity, HIV-serostatus, long-term relationship status, # male sexual partners, drug use in previous 90 days, positive and negative affect, social desirability, compulsive sexual behavior, and internalized homonegativity. e Controlling for age, ethnicity, education, and religiosity. *p < 0.05; **p < 0.01; ***p < 0.001. MSM, men who have sex with men.

a

Wu et al. (2014); CrossRepresentative Male China sectional community sample of rural-to-urban migrants (n = 4,069)

Wright (2013); United States

Wright et al. (2013); United States

Population (n)

General social survey data from 2006 and 2008 (n at Time 1 [T1] = 1,021, n at Time 2 [T2] = 1,022) CrossGeneral social survey sectional data from 23 time points(independent samples) during 1973–2010 (n = 14,193)

Cohort

Type of study

Wright (2012); United States

Author, year, and country

Table 2. (Continued)

PORNOGRAPHY USE AND SEXUAL RISK BEHAVIORS

67

Table 3. Quality Index Scale Means and Standard Deviations for Internet Pornography Studies Sexual risk behavior Nelson et al. (2014) Peter and Valkenburg (2011)

Reporting

External validity

Internal validity

Power

Total

UAI Condom use

3 7

0 2

4 4

0 0

7 13

M SD

5.00 2.83

1.00 1.41

4.00 0.00

0.00 0.00

10.00 4.24

et al.32 examined a nonmonogamous sample of MSM, and found that watching pornography containing unprotected anal intercourse was correlated with engaging in both unprotected insertive and receptive anal intercourse. Similarly, in a sample of MSM, Nelson et al.35 found that the odds of engaging in unprotected anal intercourse were higher as the amount of unprotected anal intercourse viewed in pornography increased. In addition, the odds of having knowingly engaged in serodiscordant unprotected anal intercourse were also higher for greater amounts of unprotected anal intercourse viewed in pornography. Rosser et al.33 found that MSM who reported using pornography for more than 1 hour per day were more likely to have engaged in unprotected receptive anal intercourse or HIV serodiscordant anal intercourse than those who viewed less or none. This association was, however, no longer significant when other variables, including the preference for viewing pornography with or without unprotected anal intercourse were controlled for, suggesting that this correlation may be partly explained by personal preferences, rather than the influence of pornography. Furthermore, those who had a preference for viewing pornography containing unprotected anal sex reported more unprotected anal intercourse than those who had no preference, while those with a preference for viewing pornography depicting protected anal intercourse reported significantly lower risk behavior.33 In addition, no associations for unprotected insertive intercourse were significant, which contradicted the observations of Eaton et al.23

Examining young Croatian adults in a nationwide survey, Sinkovic et al.31 created an index of sexual risk taking, which included whether participants had an early sexual debut ( < 15 years), condom use during the most recent sexual encounter, consistent condom use over the past month, previous experience of concurrent sexual relationships, or previous experience of sexual intercourse with an unknown person (casual sex). A higher sexual risk taking index was correlated with younger age of first exposure to pornography in both males and females. Despite this, frequency of pornography use over the previous 12 months was not linked to sexual risk taking for either gender. Wright34 examined the correlation between pornography and condom use across eight time points between 1996 and 2010 in males from the general population in the United States. Overall, unmarried males who consumed pornography were more likely to use condoms than unmarried nonconsumers, even after controlling for relevant demographic variables. This result contrasts with Nelson et al.,35 Eaton et al.,23 and Stein et al.,32 although this association was weak and inconsistent across time frames. In addition, this study did not differentiate participants based upon sexuality, whereas the above studies examined only MSM. The disparate result may reflect the dichotomous measure of pornography use that Wright34 utilized, as it did not take into account viewing frequency, viewing duration, history of pornography use, or how recent the use had been.

Table 4. Quality Index Scale Means and Standard Deviations for Unspecified Pornography Studies Sexual risk behavior Carroll et al. (2008) Eaton et al. (2012) Kraus and Russell (2008) Morgan (2011) Poulsen et al. (2013) Rosser et al. (2013) Sinkovic et al. (2012) Stein et al. (2012) Stulhofer et al. (2010) Traeen (2013) Weinberg et al. (2010) Wright (2012) Wright (2013) Wright et al. (2013) Wu et al. (2014)

Reporting

External validity

Internal validity

Power

# Sexual partners # Sexual partners, UAI # Sexual partners # Sexual and # casual partners # Sexual partners UAI Sexual risk taking index UAI # Sexual partners # Sexual partners # Sexual/casual partners Engagement in casual sex # Sexual partners, condom use # Sexual partners Multiple sexual partners

5 7 4 4 4 5 5 6 6 2 3 4 6 6 4

0 2 0 0 0 1 3 2 0 2 0 3 3 3 3

4 4 2 4 3 3 4 4 4 3 3 4 4 4 4

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

M SD

4.73 1.33

1.47 1.36

3.60 0.63

0.00 0.00

Total 9 13 6 8 7 9 12 12 10 7 6 11 13 13 11 9.80 2.57

68

Only one study specifically investigated the association between Internet pornography and unsafe sex practices. Peter and Valkenburg30 examined a random community sample across two time points 6 months apart. Increased frequency of Internet pornography use in the preceding 6 months was associated with casual sex without a condom at both time points, and this correlation remained significant after controlling for relevant demographic variables. It is of note that these findings represented a small effect size, and contrasted with the results of the comparative sample reported in Sinkovic et al.31 Despite this, the findings appear to be consistent with MSM samples.23,32,33,35 Number of sexual partners

Ten studies23,34,36–43 examined the association between pornography use and number of sexual partners. The overall results suggested that there is a robust association between pornography use and a higher number of sexual partners. No studies specifically examined Internet pornography in relation to this variable. Carroll et al.36 found that in a sample of university students, pornography viewing frequency was associated with a higher number of lifetime sexual partners for both genders, and a higher number of sexual partners over the previous 12 months in females but not males, although the effects sizes were very small (see Table 2). Additionally, compared to nonusers, male monthly pornography users had a significantly greater number of partners over the previous 12 months, while those who viewed it daily reported lower numbers. Similarly, Traeen and Daneback42 found that time spent viewing pornography while masturbating was significantly associated with a greater number of sexual partners over the past 5 years, 3 years, and 12 months for heterosexual but not nonheterosexual men and women. The association with lifetime number of partners was only significant for heterosexual women. In addition, time spent watching pornography while having sex was correlated with number of sexual partners in heterosexual men but not women and the number of partners over the previous 3 years for nonheterosexual women but not men. Contrasting the findings, Traeen and Daneback42 and Eaton et al.23 established that time spent viewing pornography per week was significantly associated with a higher number of sexual partners in the previous month for HIV-negative MSM. In further support of an association between number of sexual partners and pornography use, Morgan37 found that the frequency of pornography use and the number of different types or media used were significantly and positively correlated with lifetime number of sexual partners for both males and females. However, these findings still reflected small effect sizes. Supporting other findings,37,42 in a study conducted within the general population across multiple time points, pornography use was associated with a higher number of sexual partners over both the previous 12 months and 5 years for both females40 and males,34 when controlling for demographic variables. These correlations represented a small to medium effect size. However, the unique contribution of pornography use was very low, accounting for only 1–2% and 3% of the variance for females and males respectively. In concurrence with Wright34 and Wright et al.,40 Stulhofer et al.39 found that paraphilic pornography users had a significanly greater lifetime number of sexual partners than

HARKNESS ET AL.

nonparaphilic users and Poulsen et al.38 demonstrated that for both males and females, pornography users reported a significantly greater lifetime number of sexual partners than nonusers. Furthermore, Wu et al.41 found that the odds of having had multiple sexual partners were greater for higher frequencies of pornography use in Chinese rural-to-city migrant males. Only one study43 did not find an association for number of sexual partners. However, this may have been because this variable was examined in relation to exposure to pornography during adolescence and not current frequency of use. Accordingly, frequency of pornography use appears to have a robust association with the lifetime number of sexual partners, although the evidence for this association in shorter and more recent time periods is mixed and appears to be stronger for females than males. Casual sex behavior

In total, only three studies12,37,44 examined the association between pornography use and number of casual sex partners. No studies specifically examined Internet pornography in relation to this variable. Morgan37 found that in college students, the frequency of pornography use and the number of different types of media used were significantly and positively correlated with lifetime number of casual sex partners, although these findings reflected small effect sizes. Wright44 also found support for this association, establishing in a nationwide longitudinal study that pornography exposure at baseline was associated with an almost twofold increase in the odds of casual sex at a 2 year follow-up, even after controlling for baseline casual sex behavior. Conversely, Weinberg et al.12 found that for females, regardless of sexuality, frequency of pornography use was only associated with a greater number of partners in the previous 12 months with whom participants were significantly involved, and did not reflect an increase in casual sex partners. This association was not replicated for heterosexual or nonheterosexual males. Discussion

The objective of this systematic review was to evaluate the relationship between pornography consumption and indicators of sexual risk behaviors in adults. All 17 articles supported an association, with findings generally comparable across Internet pornography and unspecified pornography studies where there was available literature. Sexual risk behaviors may have serious consequences for physical and sexual health, and the findings of this review highlight the importance of this topic and the need for further research into how pornography use fits into the broader picture of sexual risk behaviors. However, there was a paucity of research, especially for condom use and casual sex behavior, and no research investigating actual history of STI incidence, which may be considered to be a more robust indicator of sexual risk behavior. Accordingly, this variable would be an important addition to future research, in order to enhance methodological rigor. The relatively low means obtained for the quality assessment suggests that methodological improvements in this field of research are needed. The main area of concern was external validity, reflecting the biased and inadequately sampled populations used in these studies. The studies often utilized convenience sampling (usually young adult college students) to recruit participants,36,37,39,45,46 and further research using

PORNOGRAPHY USE AND SEXUAL RISK BEHAVIORS

community samples would allow more generalizability of results. Random sampling and increased reporting of response rates is also recommended. In addition, many studies examined very specific populations. For example, four studies examined only MSM,23,32,33,35 and a further four studies restricted their participants to males.34,39,41,47 Poor replication does not lend confidence to the existing findings, and further replication, both within specific and broader community populations, is required. As the current review consisted entirely of predictive studies, the causal relationship between pornography use and sexual risk behaviors is unable to be determined. It may be speculated that sexual risk taking leads to pornography use or vice versa, that the effect is bidirectional, or that some other unmeasured variables are mediating this relationship. Three studies attempted to account for this: two31,43 examined age at first exposure to pornography, and one was longitudinal.44 Whilst experimental design in this area is practically and ethically difficult to conduct, if accomplished it would assist in strengthening the literature. In addition, the majority of reviewed studies did not compare indicators of sexual risk behaviors to a control population of nonpornography users, which is recommended for future work. Many studies in the review did not provide a definition, or provided inconsistent definitions, of pornography. This has been highlighted as a flaw in the literature and needs to be addressed in order for future research to be more interpretable.2 Furthermore, the measurement of pornography use was not uniform, using measures such as frequency of use, viewing time, and unreliable dichotomous variables to separate use and nonuse. As such, it is difficult to compare findings.2 Furthermore, only two studies37,39 examined how pornography genre relates to risky sexual behaviors. In both studies, support was found for differences between pornography types. Additionally, it is important to note that the studies included in this review that examined unspecified pornography use may have actually included unreported Internet pornography use. As such, it is unclear whether similarities between the results for different media types reported in this review reflect this methodological issue. Accordingly, future research should attempt to control for such factors when investigating pornography consumption and sexual risk behaviors. Finally, there was a lack of theory driven research in the literature reviewed. Only two studies39,41 attempted to test a theory. Furthermore, only four additional studies30,31,34,40,44 cited relevant theories to support their hypotheses or results. Future research would benefit from theory driven study designs, in order to clarify how pornography fits into the broader picture of sexual risk behavior. Unsafe sex practices and condom use

The reviewed literature suggests an association between pornography use and safer sex practices. However, the nature of this association was varied, and only two studies examined females.23,30–35 For MSM, findings varied between studies according to the type of intercourse, with the majority of studies only supporting an association with unprotected receptive anal intercourse. As the HIV risk for receptive anal sex has been shown to be 30 times greater than that of insertive anal sex acts,48 this may have implications for the

69

severity of sexual health risk. The findings were less consistent for heterosexual samples, with only weak support for a correlation between pornography use and tendency to practice unsafe sex. Moreover, only one study examined the association between Internet pornography consumption and unsafe sex practices.30 Thus, while the results mirror unspecified pornography studies, the available conclusions are currently limited. Number of sexual partners

The appraised literature supports a strong association between a higher number of sexual partners and tendency to use pornography.23,34,36–43 All but one of the studies included in the review found that pornography use was positively correlated with lifetime number of sexual partners. However, number of sexual partners has not been examined in relation to Internet pornography use, so the relationship between these variables is unknown. Casual sex behavior

There was a paucity of research in this area, with only three studies12,37,44 examining the relationship between pornography use and casual sex behavior. The findings were mixed and require further replication and clarification due to the greater sexual health risk of casual sex when compared to sex with regular partners. Conclusions

Taken together, the results from the current review demonstrate that there is an association between pornography use and sexually risky behaviors. Where the literature was available, the results appear to be consistent across both Internet pornography and pornography. These findings contribute to the existing body of evidence from adolescents (e.g., Braun-Courville and Rojas, Luder et al., and Wingood et al.16–18). The current review was limited in several ways, first, in the scope of behaviors examined. This did not cover attitudes that may affect or predispose risky sexual behaviors.49 Furthermore, the scope of the review was not exhaustive in terms of the indicators of sexual risk behaviors identified. In addition, comparison of Internet and unspecified pornography studies was limited due to the lack of comparable effect sizes. A meta-analysis may be appropriate for future reviews, when a greater number of studies are available. While beyond the scope of the current review, a systematic review or meta-analysis conducted on the adolescent literature would also be valuable and would allow comparison with adult findings. This area of research is still in its infancy, and future research should seek to shed light on the factors that contribute to risky sexual behaviors. Sexual risk behaviors are an important public and sexual health issue, as they may lead to increased STI/HIV transmission and poor health outcomes. While no directional inferences can be made regarding the link between sexual risk behavior and pornography consumption, in light of the current review, it may be pertinent to consider potential clinical implications. For example, the findings may contribute to informing programs to build ‘‘porn literacy,’’ assisting consumers to navigate pornography

70

HARKNESS ET AL.

effectively.50 Critical analysis of sexual media may be promoted through sex education programs, humorous videos, or instructional games.50 In addition, certain pornography users (e.g., paraphilic users, MSM) may be potential targets for sexual health interventions such as Internet interventions,9,51,52 and this area certainly warrants further investigation. Notes

a. For the purposes of the review, the term ‘‘unspecified pornography’’ is used for studies that have not clarified the form (i.e., magazines, books, videos, Internet) of pornography used, or differentiated between Internet and non-Internet pornography. Author Disclosure Statement

No competing financial interests exist. References

1. Ropelato J. (2007) Internet pornography statistics. http:// internet-filter-review.toptenreviews.com/internet-pornographystatistics.html (accessed Jun 2, 2012). 2. Short M, Black L, Smith A, et al. A review of internet pornography use research: methodology and content from the past 10 years. Cyberpsychology, Behavior, & Social Networking 2012; 15:13–23. 3. Hald GM, Malamuth NM. Self-perceived effects of pornography consumption. Archives of Sexual Behavior 2008; 37:614–625. 4. Do¨ring NM. The current state of discussion on pornography ethics: from anti-porn and anti-censorship to pro-porn positions. Zeitschrift fur Sexualforschung 2011; 24:1–30. 5. Do¨ring NM. The Internet’s impact on sexuality: a critical review of 15 years of research. Computers in Human Behavior 2009; 25:1089–1101. 6. Cooper A. Sexuality and the Internet: surfing into the new millennium. CyberPsychology & Behavior 1998; 1:187–193. 7. Byers LJ, Menzies KS, O’Grady WL. The impact of computer variables on the viewing and sending of sexually explicit material on the Internet: testing Cooper’s ‘‘TripleA Engine.’’ The Canadian Journal of Human Sexuality 2004; 13:157–169. 8. Fisher WA, Barak A. Internet pornography: a social psychological perspective on Internet sexuality. Journal of Sex Research 2001; 38:312–323. 9. Rosser BRS, Grey JA, Wilkerson JM, et al. A commentary on the role of sexually explicit media (SEM) in the transmission and prevention of HIV among men who have sex with men (MSM). AIDS & Behavior 2012; 16:1373–1381. 10. Gagnon JH, Simon W. (1973) Sexual conduct: the social origins of human sexuality. Chicago: Aldine. 11. Simon W, Gagnon JH. Sexual scripts: permanence and change. Archives of Sexual Behavior 1986; 15:97–120. 12. Weinberg MS, Williams CJ, Kleiner S, et al. Pornography, normalization, and empowerment. Archives of Sexual Behavior 2010; 39:1389–1401. 13. Wright PJ. Mass media effects on youth sexual behavior. Communication Yearbook 2011; 35:343–386. 14. Bandura A. (1986) Social foundations of thought and action: a social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

15. Seto MC, Maric A, Barbaree HE. The role of pornography in the etiology of sexual aggression. Aggression & Violent Behavior 2001; 6:35–53. 16. Braun-Courville DK, Rojas M. Exposure to sexually explicit Web sites and adolescent sexual attitudes and behaviors. Journal of Adolescent Health 2009; 45:156–162. 17. Luder M-T, Pittet I, Berchtold A, et al. Associations between online pornography and sexual behavior among adolescents: myth or reality? Archives of Sexual Behavior 2011; 40:1027–1035. 18. Wingood GM, DiClemente RJ, Harrington K, et al. Exposure to X-rated movies and adolescents’ sexual and contraceptive-related attitudes and behaviors. Pediatrics 2001; 107:1116–1119. 19. Peter J, Valkenburg PM. The use of sexually explicit Internet material and its antecedents: a longitudinal comparison of adolescents and adults. Archives of Sexual Behavior 2011; 40:1015–1025. 20. Kubicek K, Beyer WJ, Weiss G, et al. In the dark: young men’s stories of sexual initiation in the absence of relevant sexual health information. Health Education & Behavior 2010; 37:243–263. 21. Green ST. HIV and AIDS, the Internet pornography industry and safer sex. International Journal of STD & AIDS 2004; 15:206–208. 22. Paul EL, McManus B, Hayes A. ‘‘Hookups’’: characteristics and correlates of college students’ spontaneous and anonymous sexual experiences. Journal of Sex Research 2000; 37:76–88. 23. Eaton LA, Cain DN, Pope H, et al. The relationship between pornography use and sexual behaviours among atrisk HIV-negative men who have sex with men. Sexual Health 2012; 9:166–170. 24. Kelly JA, Murphy DA, Sikkema KJ, et al. Psychological interventions to prevent HIV infection are urgently needed: new priorities for behavioral research in the second decade of AIDS. American Psychologist 1993; 48:1023. 25. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. Journal of Epidemiology & Community Health 1998; 52:377–384. 26. Ferro MA, Speechley KN. Depressive symptoms among mothers of children with epilepsy: a review of prevalence, associated factors, and impact on children. Epilepsia 2009; 50:2344–2354. 27. Olivo SA, Macedo LG, Gadotti IC, et al. Scales to assess the quality of randomized controlled trials: a systematic review. Physical Therapy 2008; 88:156–175. 28. Sanderson S, Tatt ID, Higgins JP. Tools for assessing quality and susceptibility to bias in observational studies in epidemiology: a systematic review and annotated bibliography. International Journal of Epidemiology 2007; 36: 666–676. 29. Wang T, Collet J-P, Shapiro S, et al. Adverse effects of medical cannabinoids: a systematic review. Canadian Medical Association Journal 2008; 178:1669–1678. 30. Peter J, Valkenburg PM. The influence of sexually explicit internet material on sexual risk behavior: a comparison of adolescents and adults. Journal of Health Communication 2011; 16:750–765. 31. Sinkovic M, Stulhofer A, Bozic J. Revisiting the association between pornography use and risky sexual behaviors:

PORNOGRAPHY USE AND SEXUAL RISK BEHAVIORS

32.

33.

34. 35.

36.

37.

38. 39. 40.

41.

42.

43.

the role of early exposure to pornography and sexual sensation seeking. Journal of Sex Research 2013; 50:633–641. Stein D, Silvera R, Hagerty R, et al. Viewing pornography depicting unprotected anal intercourse: are there implications for HIV prevention among men who have sex with men? Archives of Sexual Behavior 2012; 41:411–419. Rosser BRS, Smolenski D, Erickson D, et al. The effects of gay sexually explicit media on the HIV risk behavior of men who have sex with men. AIDS & Behavior 2013; 17:1488–1498. Wright PJ. U.S. males and pornography, 1973–2010: consumption, predictors, correlates. Journal of Sex Research 2013; 50:60–71. Nelson KM, Simoni JM, Morrison DM, et al. Sexually explicit online media and sexual risk among men who have sex with men in the United States. Archives of Sex Behavior 2014; 43:833–843. Carroll JS, Padilla-Walker LM, Nelson LJ, et al. Generation XXX: pornography acceptance and use among emerging adults. Journal of Adolescent Research 2008; 23: 6–30. Morgan EM. Associations between young adults’ use of sexually explicit materials and their sexual preferences, behaviors, and satisfaction. Journal of Sex Research 2011; 48:520–530. Poulsen FO, Busby DM, Galovan AM. Pornography use: who uses it and how it is associated with couple outcomes. Journal of Sex Research 2012; 50:72–83. Stulhofer A, Busko V, Landripet I. Pornography, sexual socialization, and satisfaction among young men. Archives of Sexual Behavior 2010; 39:168–178. Wright PJ, Bae S, Funk M. United States women and pornography through four decades: exposure, attitudes, behaviors, individual differences. Archives of Sexual Behavior 2013; 42:1131–1144. Wu JQ, Wang KW, Zhao R, et al. Male rural-to-urban migrants and risky sexual behavior: a cross-sectional study in Shanghai, China. International Journal of Environmental Research & Public Health 2014; 11:2846–2864. Traeen B, Daneback K. The use of pornography and sexual behaviour among Norwegian men and women of differing sexual orientation. Sexologies: European Journal of Sexology & Sexual Health/Revue europeenne de sexologie et de sante sexuelle 2013; 22:e41–e8. Kraus SW, Russell B. Early sexual experiences: the role of Internet access and sexually explicit material. CyberPsychology & Behavior 2008; 11:162–168.

71

44. Wright PJ. A longitudinal analysis of US adults’ pornography exposure: sexual socialization, selective exposure, and the moderating role of unhappiness. Journal of Media Psychology: Theories, Methods, & Applications 2012; 24: 67–76. 45. Perry M, Accordino MP, Hewes RL. An investigation of Internet use, sexual and nonsexual sensation seeking, and sexual compulsivity among college students. Sexual Addiction & Compulsivity 2007; 14:321–335. 46. Sˇtulhofer A, Jelovica V, Ruzic´ J. Is early exposure to pornography a risk factor for sexual compulsivity? Findings from an online survey among young heterosexual adults. International Journal of Sexual Health 2008; 20:270–280. 47. Twohig MP, Crosby JM, Cox JM. Viewing Internet pornography: for whom is it problematic, how, and why? Sexual Addiction & Compulsivity 2009; 16:253–266. 48. Australian National Council on AIDS Hepatitis C and Related Diseases. (2001) Guidelines for the management and post exposure prophylaxis of individuals who sustain nonoccupational exposure to HIV. The ANCHARD Bulletin, p. 28. 49. Conley TD, Collins BE. Differences between condom users and condom nonusers in their multidimensional condom attitudes. Journal of Applied Social Psychology 2005; 35: 603–620. 50. Albury K. Porn and sex education, porn as sex education. Porn Studies 2014; 1:172–181. 51. Rosser BRS, Wilkerson JM, Smolenski DJ, et al. The future of Internet-based HIV prevention: a report on key findings from the Men’s INTernet (MINTS-I, II) Sex Studies. AIDS & Behavior 2011; 15:91–100. 52. Hooper S, Rosser BRS, Horvath KJ, et al. An online needs assessment of a virtual community: what men who use the internet to seek sex with men want in Internet-based HIV prevention. AIDS & Behavior 2008; 12:867–875.

Address correspondence to: Dr. Barbara Mullan School of Psychology and Speech Pathology Curtin University Kent Street Bentley WA 6102 Australia E-mail: [email protected]

Copyright of CyberPsychology, Behavior & Social Networking is the property of Mary Ann Liebert, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.