Psychosocial and Cultural Processes Underlying the Epidemiological Paradox within U.S. Latino Sexual Risk: A Systematic Review

According to the epidemiological paradox, less acculturated Latina/o youth display fewer sexual risk behaviors. A systematic review was performed on psychosocial and cultural mechanisms potentially underlying the epidemiological paradox in sexual risk behaviors of U.S. Latina/o youth across acculturation measures (between January 2000 to October 2022). Thirty-five publications (n = 35) with forty-eight analyses of underlying mechanisms met the inclusion criteria. Thirty-one results from twenty-three publications found supporting evidence that one of the five factors was an underlying mechanism in the epidemiological paradox (n = 13 parenting practices, n = 4 peer influences, n = 4 familismo values, n = 4 religiosity, n = 6 traditional gender norms) as, generally protective, mediators or moderators in the link between acculturation and sexual risk behaviors. Studies varied in the sexual risk behavior examined and measurement of acculturation, but primarily employed cross-sectional designs and recruited samples through schools. Mechanisms that enhance close ties and unity of the family, such as those of familismo values and positive parenting, reduce the likelihood of sexual risk behaviors as Latina/o youth become more acculturated. Future directions are discussed which may provide guidance for risk prevention and intervention.


Introduction
The United States (U.S.) is seeing unprecedented changes in the diversity of its population, particularly in national origin, ethnicity, and language use of its youth. As the largest non-White ethnic group in the U.S. [1], Latina/os comprise the largest and fastest growing minority group of children in the U.S. with an exponential growth in the last two decades across the country [2]. Disease and disability among such a large population can have major public health consequences such that issues affecting the healthy development of Latina/o children should be of national importance [3]. To support healthy development and outcomes among this population, we must understand both distinct cultural and common social influences on health, both to understand etiological processes and identify where interventions should be targeted.

Sexual Risk Behaviors among Latina/o Youth
Because Latina/o adolescents' birth rates are the highest of any major racial/ethnic group in the U.S. [4] and their sexually transmitted infection (STI) prevalence is more than twice that of Whites [5][6][7], their sexual risk behaviors have been deemed a major public health concern [8]. Preventing or delaying the initiation of sexual risk behaviors in adolescence, therefore, can have significant public health implications. Research is needed to identify processes that influence the initiation of health risk behaviors in this developmental paradox in sexual risk behaviors for Latina/o youth can be aided through a focus on intrapersonal and interpersonal-level psychosocial and cultural processes that have been found to change as acculturation occurs.

Proposed Underlying Mechanisms of the Epidemiological Paradox
Psychosocial and cultural factors may help us understand why we find paradoxical outcomes for health risk behaviors, including sexual risk behaviors. For example, social and cultural characteristics have been found to mediate these associations [31]. Moreover, influences on adolescent health risk behaviors can range from intrapersonal, to interpersonal, to broader contextual factors according to the socioecological model of human development [32]. In fact, according to studies guided by the theory of planned behaviors [33], which emphasizes the role of attitudes, subjective peer norms, and perceived behavioral control in predicting an individual's intentions to engage in a behavior, psychosocial and cultural factors such as parenting practices and peer associations are important influences on decision-making processes which predict health behaviors [34,35]. We propose a focus on intrapersonal and interpersonal social and cultural features, which are the most proximal level of influences on behaviors and have also been examined in the majority of studies on this topic.
Social processes. A large body of work has emphasized the influence of psychosocial processes on health risk behaviors that are found across cultures and contexts. Two domains of psychosocial influences on adolescent health risk behaviors including parenting features (i.e., monitoring, involvement, and communication) and peer influences (i.e., peer norms and modeling). These parental and peer factors may underlie paradoxical findings among Latina/os by both influencing sexual risk behaviors as well as change across acculturation.
Parental influences. Parental influence encompasses a wide range of attitudes or actions (e.g., monitoring, communication) that somehow shapes or molds the child's attitudes or behaviors. Many parenting practices may serve as protective influences against engaging in risk behaviors among youth. When adolescents perceive low levels of key parental behaviors, such as monitoring, they are more likely to engage in various health risk behaviors [36][37][38]. Additionally, parenting practices can also vary with acculturation. In fact, parenting practices, such as permissiveness, change among Latinos with each successive generation in the U.S. with an increase in proportion of permissive parents [39]. Subsequently, parenting influences may explain differences in sexual risk behaviors associated with acculturation.
Peer influences. Similarly, to parental influences, peer factors vary according to acculturation and, in turn, influence sexual risk behaviors. Peer influences appear to exert the ability to influence individual behavior among members of a group based on group norms. Specifically, peer norms, or perceiving that peers are engaging in a risk behavior, have been associated with the initiation and engagement of various health risk behaviors [40][41][42]. Moreover, Latina/o adolescents whose families were more recent migrants to the U.S. have been found to be more resistant to peer pressure than those of later generations [43]. Although important, these typical social processes of parental and peer influences do not account for the unique cultural context of Latina/o youth that may contribute to the epidemiological paradox.
Cultural processes. One hypothesis posited to explain the epidemiological paradox includes the erosion of protective cultural processes as acculturation occurs. Research suggests that certain features of the Latino culture may enhance resilience, such as familism and religiousness [44]. Specifically, the retention of some traditional cultural values appears to protect adolescents from engaging in health risk behaviors [45]. However, some cultural factors such as traditional gender norms, particularly among males, increase engagement in health risk behaviors such as violence and sexual risk behaviors [46]. It is therefore important that cultural factors from intrapersonal to interpersonal levels of influence be examined as possible underlying mechanisms of the paradox. Familism, religiosity, and traditional gender norms have been the most highly examined cultural factors in this context.
Familismo. Strong family traditions anchor the upbringing of many Latina/o children. Furthermore, the cultural value of familismo, which places focus on family as a collective unit while emphasizing family bonding, dependence, obligation, and support [47], reduces the likelihood of sexual risk behavior among Latina/o youth [10,48,49]. However, more acculturated families are less likely to hold onto values and characteristics associated with familismo [14,27,47,50]. In fact, first-generation Latinos have the highest levels of family cohesion [51].
Religiosity. Within the Latino culture, religiosity is a pervasive force, guiding attitudes, behaviors, and even social interactions for many [52][53][54][55]. Religiosity, referring to the quality or state of religious beliefs or practices, may encompass the endorsement of moral values that guide an individual's behaviors. In fact, religiosity has a protective association with various sexual risk behaviors including delayed sexual debut and fewer sexual partners [56]. Moreover, less acculturated Latina/os use religious coping strategies more frequently than those with higher levels of acculturation [57]. This deterioration of protective religious dimensions among more acculturated Latina/os may account for the epidemiological paradox in sexual risk outcomes observed among Latina/o youth.
Traditional gender norms. Traditional gender norms among Latina/o males and females are often attached to sexual values including notions of female virginity, sexual desire, and sexual communication, as well as affect sexual behaviors [58,59]. Specifically, machismo, whereby males are expected to adhere to a heightened masculine role and have little control over sexual impulses, has been associated with unprotected sex and more sexual partners among Latino adult males [60]. On the other hand, ascribing to marianismo, a traditional aspect of the female gender role that emphasizes virtues such as purity and moral strength, has been associated with reduced sexual activity among females [61]. However, inconsistent condom use has been associated with both of these gender norms among Latina/o youth [58]. Moreover, as acculturation levels increase, Latina/o individuals replace their traditional view and practice of gender roles with those of the mainstream U.S. culture [50,62]. Overall, the influence of cultural factors on sexual behaviors across acculturation processes is complex but may help us better unravel the epidemiological paradox.

Aims
Thus far, no review has examined the influence of both social processes that cross cultural bounds and Latino cultural factors as underlying mechanisms of the epidemiological paradox for sexual risk behaviors observed among U.S. Latina/o youth. Although there is research supporting that psychosocial and cultural factors operate as underlying mechanisms of the paradox, no theories or implications have been reached to illuminate why such factors explain paradoxical findings in this area which could aid in prevention strategies. Additionally, method strategies employed in such studies including the measurements used to examine the epidemiological paradox, such as generational status, language proficiency, and ethnic identification, have not been critically reviewed in this body of research. Moreover, consensus has not been established whether some cultural factors, such as traditional gender norms, lend to resiliency or risk for sexual risk behaviors among Latina/o youth.
Thus, this systematic review will (1) synthesize findings of intrapersonal and interpersonal social (i.e., parental and peer influences) and cultural (i.e., familismo, religiosity, and traditional gender norms) factors that have been found to moderate and/or mediate the epidemiological paradox for sexual risk behaviors among U.S. Latina/o youth in that the measure of acculturation will interact significantly with the sociocultural variables to increase or decrease sexual risk outcomes; (2) critique this body of research; (3) propose future directions; and (4) draw implications from this research.

Search Strategy
The search strategy employed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A search of Google Scholar, PubMed, PsychINFO, EBSCO, and reference lists of eligible papers was conducted using the following key terms: Latino adolescent sexual risk behaviors, parental influence, peer influence, generation status, acculturation, (English) language use/preference, epidemiological paradox, familismo, gender norms, marianismo, machismo, and religiosity. A sequential process of examining the title, abstract, and main text content of each article was undertaken, with exclusion of articles occurring at each stage.

Eligibility Criteria
Studies were selected for inclusion according to the following criteria: (1) quantitative empirical publications from peer-reviewed journals published between January 2000-October 2022 in English or Spanish; (2) at least one intrapersonal or interpersonal social (i.e., parental and peer influences) and/or cultural factor (i.e., familismo, religiosity, and traditional gender norms) was included in the analysis; (3) a measurement of the epidemiological paradox through indicators of acculturation were included in the analysis; and (4) outcomes examined were sexual risk behaviors among U.S. Latino/Hispanic youth. Interrater reliability of Cohen's Kappa was used to assess agreement for the inclusion of studies among two coders. Figure 1 delineates the systematic search process. A total of 5362 citations were identified. An additional five studies were identified through reference lists. After screening of the title, abstract, and content for relevance, 5327 were excluded. Of the remaining 35 studies, which included 48 analyses of possible underlying mechanisms, 27 analyses examined social processes (n = 23 parental influences; n = 4 peer influences) and 21 examined cultural factors (n = 8 familismo; n = 7 religiosity; n = 6 traditional gender norms) as moderators or mediators of the epidemiological paradox in sexual risk outcomes among Latina/o youth. All included studies were published in English. Interrater reliability across the factors was high (κ = 0.81-0.97). Identification and coding of included studies was based on two raters for each factor. Each coder extracted information, included in Table 1, and calculated their agreement based on Cohen's Kappa for extracted study information. About one-fourth of the included studies examined more than one factor (n = 13). The majority of the included studies (n = 30, over 85%) found significant mediating or moderating effects of sociocultural factors in sexual risk behaviors among Latina/o adolescents. Moreover, most studies that found significant effects examined two types of sexual risk behaviors:

Synthesis of Review
(1) use of barrier methods, and (2) initiation of sexual intercourse including intentions. Table 1 details the key characteristics and results of the included publications for each factor. Descriptions regarding the measures of acculturation used within this set of studies are provided in Table 2.   Caregiver-adolescent acculturation discrepancies (measured by the BIQ-S b ).

•
Higher levels of caregiver family functioning had significant effects on inconsistent condom use (β = 0.13).

•
No significant indirect effects were found between acculturation components and reports of condom use through family functioning. Generational status (youth and parent nativity) and language acculturation (language spoken at home)

•
Birth control recommendation (BCR) was positively related to sexual intercourse for second-generation youth (β = 0.48).

•
Among first-generation females higher BCR decreased the probability of sexual intercourse 1 year later (OR = 0.22), whereas among second-generation girls, higher BCR increased the probability of subsequent sexual intercourse (OR = 0.63).

•
For second-generation youth only, maternal BCR was significantly associated with increased condom use (β = 0.70).

•
Higher levels of BCR were related to a higher probability of using condoms for second-generation girls and lower probabilities of using condoms for third-generation girls (OR = 0.29). Greater monitoring of location was associated with less risky sexual behavior, but only for youth second-generation and above (β = −1.05). • Parental monitoring did not mediate the relationship between immigrant generation and either of the two partner risks for which significant differences by generation were found (i.e., partner used alcohol/marijuana frequently and partner gang affiliated/incarcerated). • Nonetheless, parental monitoring maintained an independent relationship with partner risk: stronger parental monitoring was associated with decreased odds of choosing higher-risk partners (OR = 0.07). • Cross-sectional study.
Parent and adolescent language acculturation (measured by the SASH d ).

•
When mother and adolescent were both high in acculturation (M = 3.56), adolescents reported higher degrees of sexual experience than when the mother was low and adolescent high in acculturation (M = 2.24; F = 4.05). • Yet, mother-adolescent acculturation group was not associated with mother-adolescent conflict and could not be considered a mediator. Americanism and Hispanicism (measured by the BIQ g ).
• Assimilated adolescents reported the poorest family functioning, but adolescent assimilation negatively predicted adolescent sexual activity and unprotected sex indirectly through family functioning (β = −0.07). • In families where adolescents perceived family functioning to be more positive than parents did, adolescents appeared to be at greatest risk for sexual activity (β = 0.27) and unprotected sex (β = 0.27). However, acculturation did not predict this discrepancy in reports of family functioning. Immigration status (first and second) and primary language at home.
Neither immigration status nor acculturation moderated the link between maternal parenting constructs and risky sexual behaviors. Caregiver-adolescent acculturation discrepancies (i.e., cultural practices, values, and identities).
• Higher levels of caregiver family functioning had significant effects on inconsistent condom use (β = 0.13). • However, no statistically significant indirect effects were found between acculturation components and reports of condom use through family functioning.   ARSMA-II f .  • Number of sexual partners was positively related to time in the US (r = 0. 33) but was negatively associated with positive religious coping (r = −0.16). • Sex under the influence of alcohol was positively associated with time in the US (r = 0.10) but was not related to religious coping. Nativity, childhood and current language, and ethnic identity (using the MEIM h ).

•
In addition, considering satisfaction of sexual needs important (machismo) was associated with more sexual partners among those who attached little value to female virginity (OR = 0.7). • For males, the importance of satisfying sexual needs increased with the number of lifetime (OR = 1.4), recent sexual partners (OR = 1.1), and with inconsistent condom use (OR = 1.9).    Note: Ten studies are included in more than one section of the table because they examined more than one social and/or cultural factor. * OR = Odds ratio; HR = Hazards ratio; RRR = Relative risk ratio. a Bidimensional Acculturation Scale [13]. b Bicultural Involvement Questionnaire-Short Version (BIQ-S) [96]. c Bicultural Involvement Questionnaire-Revised [97]. d Short Acculturation Scale for Hispanics [98]. e Short Acculturation Scale for Hispanic Youth [99]. f Acculturation rating scale for Mexican Americans -II [14]. g Bicultural Involvement Questionnaire [100]. h Multi-Ethnic Identity Measure [101]. i Stephenson Multigroup Acculturation Scale (SMAS) [102]. Table 2. Summary of acculturation scales used in included studies.

Acculturation Scale # of Items (Subscales) Description
a Bidimensional Acculturation Scale [13]. 24 (2) Measures two major dimensions of acculturation (Hispanic and non-Hispanic) using 12 items (per cultural domain) measuring 3 language-related areas (English media use, language use, and proficiency).
42 (2) Measures the level of orientation toward American (21 items) and Hispanic (21 items) cultures in terms of both (a) comfort with and enjoyment of American and Hispanic cultural practices (e.g., comfort and use of language, food, and traditions) and (b) how much participants would want or like to utilize American and Hispanic cultural practices. d Short Acculturation Scale for Hispanics [98]. f Acculturation rating scale for Mexican Americans -II (ARSMA-II) [14].

(2)
Measures behavioral and affective aspects of acculturation. Two subscales (measuring integration and assimilation, as well as marginalization and separation) capturing orientations to Anglo (13 items) and Mexican (17 items Measures language use and involvement in both Latino and mainstream American activities. It yields two sets of scores to derive a measure of bicultural involvement, with individuals who are highly involved in both cultures scoring highest on the scale. A total of 21 items assess U.S. practices (e.g., speaking English, eating American food, associating with American friends), and the other 21 items assess Hispanic practices (e.g., speaking Spanish, eating Hispanic food, associating with Hispanic friends).
h Multi-Ethnic Identity Measure [101]. 15 (2) Measures ethnic identity based on the elements that are common across groups so that it can be used with all ethnic groups. Comprised of 2 factors, ethnic identity search (5 items; a developmental and cognitive component) and affirmation, belonging, and commitment (7 items; an affective component). Three items are used only for the purposes of identification and categorization by ethnicity. i Stephenson Multigroup Acculturation Scale (SMAS) [102].

(2)
Measures the strength of practices in one's heritage culture and U.S. culture. Created to measure engagement in cultural practices among members of any ethnic group and not one specific group. Exploratory factor analyses generated a 2-factor solution from the 32-item questionnaire: ethnic society immersion and dominant society immersion.
Note: Superscript notation correspond to notation of position in Table 1.
Specifically, positive parenting practices, such as monitoring and communication, were protective against risky sexual activity in that adolescents with lower acculturation and greater positive parenting practices were approximately half as likely as those of higher acculturation and lower positive parenting behaviors to engage in sexual risk behaviors. In fact, parental acculturation also predicted adolescent sexual risk behaviors and was partially mediated by parenting practices [79]. Moreover, parental influences had an indirect effect on adolescent sexual risk behaviors through dating behaviors, such that greater acculturation among females was associated with a perceived lower maternal approval of dating which was associated with a lower likelihood of being in a relationship, which, in turn, predicted lower intentions to engage in sex in the future [68].

Peer Influences
Summary. Four studies examined peer influences as underlying mechanisms of the epidemiological paradox in Latina/o adolescent sexual risk behaviors. One employed a longitudinal design [64] and all others employed cross-sectional designs [69,75,79]. Outcomes examined included sexual behaviors by three studies [64,69,79] and sexual intentions by one study [70]. One study used language as a proxy of acculturation [69], two used nativity and/or generational status [64,75], and one used psychometric scales to measure acculturation discrepancies between parent and child [79]. Peer influences examined within the selected studies included peer pressure [69], deviant peer affiliations [75], and perceived peer sexual behavior [64,79].
Results. Three of the studies found that peer influences partially mediated associations between acculturation and sexual risk behaviors [64,75,79]. Specifically, perceived peer sexual behavior had a negative and low association to sexual risk behaviors by way of an acculturation gap between parents and adolescent children [79]. Moreover, deviant peer affiliations were a secondary mediation pathway between acculturation and sexual risk behaviors by way of parental influences. In fact, among greater acculturated Latina/o youth, lower paternal acceptance and disclosure to mothers were related to greater deviant peer affiliations, and greater deviant peer affiliations were moderately linked to greater intentions for sex [75].
Results. Three of the studies did not support familismo as an underlying mechanism between associations of acculturation and sexual risk behaviors among Latina/o youth [65,76,85,87] in that associations between acculturation and/or familismo or sexual outcomes were not significant. However, four studies found that familismo partially mediated associations between acculturation and some sexual behaviors by its own moderate associations with each variable [66,68,86,88]. In fact, familismo values were protective against risky sexual activity so that adolescents with higher acculturation and less familismo values were almost twice as likely as those of lower acculturation to engage in sexual risk behaviors. Moreover, familismo had an indirect effect on sexual risk behaviors through parental and peer influences, such that greater acculturation among males was associated with a lower preference of a romantic partner's embracement of familismo, which, in turn, was associated with greater intentions to engage in sex in the future [68].

Religiosity
Summary. Seven studies examined religiosity as an explanatory mechanism [77,[89][90][91][92][93][94], six of which were cross-sectional and one longitudinal [92]. Whereas five of the studies examined sexual activity and condom use [89][90][91][92]94], one examined knowledge about condom use [77], and one examined both voluntary and involuntary sexual activity as an outcome [93]. Two of the studies included religious affiliation along with a measurement of religiosity [77,92] and one examined positive religious coping [91]. Six of the studies examined some form of linguistic acculturation, one additionally included a measure of nativity [89], one included a validated measure of acculturation [91], and two included the length of time in the U.S. [91,92].
Results. Four of the studies found religiosity to be a significant and moderate mediator between acculturation and sexual risk behaviors [89,90,93,94]. Specifically, religiosity appears to fully mediate this association as a protective factor [93] in which less acculturated adolescents with higher religiosity were less likely to engage in various sexual risk behaviors [90]. However, the type of religiosity is associated with whether it is protective or risk-enhancing, such that intrinsic religiosity is protective while extrinsic increases the likelihood of engaging in risk behaviors [94].

Traditional Gender Norms
Summary. Six studies were retrieved that examined traditional gender norm values of marianismo and/or machismo [58,61,85,86,91,95]. Five were cross-sectional studies [58,61,85,91,95] and one was a retrospective longitudinal study [86]. Five of the studies examined either sexual behaviors and/or condom use, and one study examined attitudes toward condom use and intentions to use condoms in the future [85]. Two studies used validated scales of acculturation [85,91], all other studies used a measure of linguistic acculturation, and one study additionally included measurement of nativity status [86]. Various values of marianismo and machismo were examined including importance of female virginity or chastity [58,86,91,95], family pillar [91], importance of satisfying sexual needs [58], considering sexual talk disrespectful [95], and gender role orientation [61,85].
Results. All studies demonstrated support for partial mediation by gender role norms of the link between acculturation and sexual risk behaviors. Marianismo and machismo values were found to have moderate to low associations with acculturation and delayed engagement in sexual behaviors, particularly among females who regard female virginity as important [58,86,91,95]. Specifically, the lower the acculturation level and the more traditional the gender role orientation, the greater a delay in initiating sexual intercourse [61].

Discussion
A total of 35 publications, with 48 analyses, evaluated 1 or more of the social or cultural factors that were hypothesized to explain the epidemiological paradox in sexual risk behaviors among Latina/o youth. Thirteen of these studies examined more than one of the underlying factors of interest. Longitudinal designs were employed in 17 studies. Twenty-seven results from the thirty-five studies supported one or more of the social and cultural factors as underlying mechanisms of the epidemiological paradox, with low to moderate associations between acculturation measures, underlying mechanisms, and sexual risk behavior outcomes. The underlying mechanisms examined within these studies were generally found to be protective against engaging in sexual risk behaviors, with the exception of negative peer influences.
Yet, 18 analyses from 13 studies reported non-significant findings. Discrepant findings are likely due to the large variation in how acculturation has been measured and underrepresentation of several factors of interest within empirical studies, particularly of peer influences and traditional gender norms. Consequently, no firm conclusions should yet be drawn about whether some of the hypothesized underlying mechanisms explain the epidemiological paradox in Latina/o youths' sexual risk behaviors. However, some interpretations and recommendations for future studies can be made from the current literature.

Protective Parental and Familial Explanatory Mechanisms
Of all the underlying mechanisms reviewed here, parental influences and familismo values were most often examined. Moreover, twelve studies supported parental influences and four supported familismo as underlying mechanisms of the epidemiological paradox within Latina/o youths' sexual risk behaviors. Their protective roles in preventing sexual risk engagement may emphasize the importance of parents and families among Latina/o adolescents' everyday life as they acculturate to U.S. norms. For example, Latina/o children are more likely than children in other racial/ethnic groups to eat dinner with their families six or seven nights a week [7]. However, Latino families are often faced with challenges including economic hardship, discrimination, and neighborhood context often characterized by high crime rates and unstable housing. These challenging contextual features along with parent-child acculturation discrepancies can also influence the parenting practices employed by Latino parents and family cohesion [103,104]. Thus, strengthening positive parenting practices and family cohesion may result in positive outcomes among Latina/o adolescents.
Mechanisms that reflect close familial ties are the most clearly supported in the literature to reduce the likelihood of sexual risk behaviors. For example, familismo values are expected to reduce the risk for negative behaviors by cementing strong bonds of attachment to the family, ensuring that the family remains a strong source of influence, and fostering conventional ties that discourage Latina/o youth from engaging in a variety of problem behaviors [62,105,106]. Moreover, family support can mitigate environmental influences such as poor neighborhood quality on health risk behaviors [107] by improving adolescent social-emotional competencies [108]. Moreover, familismo values may enhance and provide a context for parents to engage in a broader range of positive parenting practices such as monitoring, involvement, and communication. Some studies in this review examined constructs of both parental and familismo influences [65,66] within a single latent variable, possibly reflecting the overlapping role that parental and familismo mechanisms share. In fact, familismo values may ensure that parents continue to play an important role in their child's behavior well into adolescence.

Growing Support for Additional Underlying Mechanisms
Although research on peer influences, religiosity, and traditional gender norms as underlying mechanisms is relatively small at this point, some preliminary interpretations can be made based on the limited number of studies available. Whereas there is evidence suggesting that they influence the association between acculturation and sexual risk behav-iors among Latina/o adolescents, for some of these factors, it is unclear whether they are strictly protective or risk-enhancing.
Cultural factors, aside from familismo, that appear to reduce the likelihood of sexual risk behavior engagement include, among others, religiosity. When controlling for education and socioeconomic status, Latinos use religious coping mechanisms more frequently than their non-Latino White counterparts [109]. Given its prominent role in Latino culture overall, it appears probable that religiosity may be influential during difficult life transitions, such as during the immigration process [110]. Studies examining religious affiliation suggest that Latina/o adolescents who identify with Christian denominations feel that religion impacts family relationships; yet adolescents who come from families in which both parents are not present express that religion exerts a negative influence [111]. Religious values are important within traditional Latino culture and, when maintained, are likely to increase conformity, diminish involvement with delinquent peers, and inhibit participation in deviant activities [112,113]. Moreover, religiosity may enhance the protective influence of parents and family cohesion such that parents may use religion to teach values and exert social control within the family context [114]. In fact, recent research suggests that parental monitoring may mediate the relationship between family religiosity and some sexual risk behavior [115].
Alternative to parental influences and familismo values, peers can exert an influence in greater odds of an adolescent engaging in health risk behaviors including sexual behaviors [42,64,116,117]. This is particularly important during adolescence as peers increase in importance and become as or more influential than parents [118,119]. Specifically, adolescents in general are influenced by perceptions of their friends' engagement in sexual behaviors and are more likely to engage in these behaviors to feel like they fit in [120]. Yet, adolescents who have conservative sexual attitudes engage less frequently in sexual behaviors, even if they perceive their peers to be engaging in risky sexual behaviors [121,122]. The same patterns of peer and friend impact apply to Latina/o adolescents, but with one noted difference. Latina teens are particularly influenced by cultural norms, which tend to be more conservative regarding sexuality [123]. Therefore, it can be posited that Latina early adolescents are likely to be more conservative and place greater weight on familial ties over friendship, which suggests that for these girls, peer behavior may not be as influential in their decisions to engage in sexual activities [123]. In fact, studies included in this review demonstrated that, when acculturation is taken into consideration, the weight of peer influences, although still significant, is reduced. Together, strong ties and a shared culture that opposes negative behaviors can lead to reduced negative outcomes among immigrants [31].
Gender role socialization within a Latino family context is influenced by the cultural concepts of marianismo for females and machismo for males [50]. Additionally, some facets of traditional gender norm values such as the importance of female virginity and satisfaction of sexual needs were associated with a sexual risk behavior in a protective or risk-enhancing way depending on the specific sexual behavior examined [58]. However, acculturation sways the influence of traditional gender roles on sexual risk behaviors. Specifically, among less acculturated females, values of marianismo may lend to a less assertive role within romantic relationships, which may raise concern about Latinas' ability to communicate regarding sexuality such as condom use [95,124].

Methodological Critique and Limitations
This research field is relatively new and challenging. Serious methodological problems were identified in various studies. First, we found an over-representation of cross-sectional designs. Twenty-seven of the selected studies employed cross-sectional designs. However, psychosocial and cultural factors can have early and long-lasting influence on an individual's behavior prior to the initiation of most risky behaviors [37]. Longitudinal designs should aid our understanding of early social and cultural factors that influence subsequent risk behaviors. Moreover, examining protective and risk-enhancing factors before children enter adolescence can better inform about critical periods in which interventions can be introduced prior to the initiation of most health risk behaviors.
Second, a broad variation of acculturation measurements has been used. Whereas some studies utilized language as an acculturation proxy, others have depended on nativity or generation status, and yet others relied on validated acculturation scales. However, non-uniform use of acculturation measures raises the question about whether different studies are capturing similar aspects of acculturation. Because of their reliance on unidimensional conceptions of acculturation, most studies are limited in that it is not clear whether the epidemiological paradox is due to immigrants' acquisition of receiving-culture practices, loss of heritage-culture practices, or both [22]. In fact, some researchers argue that language use or preference measures share only small amounts of variance with more comprehensive measures and may actually capture different aspects of acculturation [125]. Alternatively, there is strong empirical evidence that supports the use of multidimensional models of acculturation over unidimensional approaches [126,127]. Some have even suggested that examining the epidemiological paradox through multidimensional constructs of acculturation may help researchers better understand underlying mechanisms of the epidemiological paradox [22]. Moreover, multidimensional constructs of acculturation should be used to examine whether biculturalism (reflecting the adoption of the receiving culture while retaining the heritage culture) [128] is the most adaptive approach to acculturation as it has been linked to better outcomes, especially among young immigrants [129,130].
Third, sample recruitment procedures evidence several limitations. Specifically, most studies relied on recruitment from educational institutions. However, due to the higher high-school drop-out rates of Latina/o in comparison to African American and White youth [131], school-based samples may not be representative of Latina/o youth across the U.S, including those at highest risk for sexual risk behaviors. Additionally, a lack of information among many studies regarding the country of origin for each study sample was notable. This is an important part of the context for understanding the results. However, the majority of studies described their sample with general terms of Hispanic or Latina/o as these populations are often treated as homogeneous. Moreover, among those that included country of origin information for their sample, most were predominantly made up of Latina/os from Mexican origins. Yet, studies have produced different results in adult samples when considering the epidemiological paradox in various health outcomes dependent on the country of origin (e.g., Puerto Rico, Mexico, Cuban) [132]. This may be due to differences in perceived discrimination, reasons for migrating, and context of reception, referring to immigrants' opportunities in the U.S. For example, Mexicans and Puerto Ricans are more likely to be marginalized, whereas Cubans generally fare better [133]. Unlike Mexicans, many of whom are undocumented and seek "under-thetable" positions [134], and Puerto Ricans, many of whom migrate to the northeast and south to escape poverty [135], many Cubans arrive in the U.S. as political refugees (although some do immigrate to escape poverty). The effects of these differences on health risk behaviors in adolescence is not well known. Multisite studies of acculturation and health outcomes are important because acculturation may take different forms depending on the context to which individuals are acculturating [136]. This approach can also aid in capturing greater diversity in Latino group samples, which should be examined as distinct country of origin groups to determine if underlying mechanisms differ by such groups.
Finally, a lack of emphasis on gender differences was evident among some reviewed studies. Despite well-documented gender differences in sexual risk behaviors as well as possible differences in social and cultural processes, some studies did not provide separate analysis of these associations for males and females or lacked consideration for one gender (i.e., exclusive female sample). For example, girls are often more highly monitored by parents and family members than boys [137,138]. Such differences in underlying mechanisms of the epidemiological paradox may further explain differences in significant findings between gender [68,74,75,88]. Analysis of underlying mechanisms between gender would help us better understand whether there are gender differences in their influence as a protective or risk factor and therefore develop better interventions specific to each gender.

Future Directions
This review suggests several important future directions in this research field. First, additional studies are required to clarify the role of each social and cultural mechanism within the epidemiological paradox of Latina/o adolescent sexual risk behaviors. Specifically, much of the current literature has focused on adult populations or other health risk behaviors among adolescents, particularly substance use. Future studies should streamline the use of acculturation measures, with a preference for multidimensional constructs of acculturation measured by psychometric scales. Subsequently, with improved methodological use of acculturation measurements and further uniformity on the operationalization of psychosocial and cultural factors, meta-analytic techniques should be employed to calculate the overall effect size of these associations.
Pubertal development plays a pivotal role in an adolescent's sexual risk behavior engagement [139]. Moreover, how and when parents discuss puberty and development with their children may influence how adolescents subsequently handle decisions of sexual behavior [140]. Such discussions may be themselves influenced by acculturation. Moreover, whether parents even discuss puberty with both boys and girls is questionable [141]. In addition, parents' timing (i.e., before the initiation of sexual behaviors) and the content of pubertal development discussions with their children are important aspects to sexual communication [140]. Therefore, a better understanding is needed of how acculturation can influence parents' approach and timing to discussions of pubertal development and sexual behaviors with their children and whether this differs between genders.
Decision-making is also an important aspect of engaging in risky behaviors. For example, the theory of planned behavior [33] posits that sexual behaviors involve cognitive processes by which an individual takes a step in deciding whether or not to act on a behavior. In fact, an adolescent's intentions have been found to be highly associated with whether they will engage in a behavior or not [142,143]. Moreover, psychosocial and cultural factors appear to exert influence on the decision-making process of adolescents' sexual behaviors across acculturation status [68,74,75]. However, it is unclear why intentions may differ across acculturation levels and how this may be influenced by social and cultural factors. Future studies should expand our understanding of social-cultural mechanisms within cognitive processes of sexual decision-making. Although some of the studies included in this review examined intentions to engage in sex as an outcome, this should be further expanded to examine the factors that influence differences in these intentions across acculturation.
For many Latino families, migration may be a fluid, unexpected, and even temporary experience. The length and timing of migration for Latinos can differ greatly. It is also important to note that in today's political climate, deportation back to the country of origin can become a reality for both parents and children. Despite being largely ignored within the forced migration literature, it has been argued that deportation is a form of forced migration that warrants attention [144]. The stress of migration processes and fear of deportation may further influence risky behaviors, as seen among Latina/o adults' drug use, HIV testing, and HIV prevalence in Mexico [145,146]. These stressors may also influence how parents approach their parenting strategies as well as family cohesion [108]. In fact, parenting practices partially mediated the relation between mothers' age at arrival and young children's social development, particularly in the case of mothers who arrived as adults [147]. Future studies should examine the role of, not just acculturation, but the migration process undertaken by Latino families when examining the epidemiological paradox. Because features of the migration are shared across groups migrating from different nations, these associations can possibly be examined among various migrant groups.
This field currently lacks a comprehensive model or theory which incorporates the role of acculturation as well as social and cultural factors to explain sexual behaviors of Latina/o youth. Some studies may use previously established models and theories to guide research examining the epidemiological paradox. However, this phenomenon is unique due to the nuanced influence of acculturation on health outcomes and its interaction with sociocultural processes, as well as its specific application to certain ethnic minority groups. Moreover, it is difficult to understand whether and how each underlying mechanism may interact with one another. Although work in this field is still growing, current findings may help the development of a theoretical model that is aimed at explaining the epidemiological paradox in health risk behaviors. Future research should have, as one aim, to support the development of a comprehensive model that accounts for both social and cultural factors across various levels of contexts within an ecological framework. Such a model would piece together the role of each social and cultural factor as well as their interaction with one another within the epidemiological paradox. Moreover, a multilevel model that accounts for broader contexts may help us understand the role of other factors, from cognitive processes to neighborhood characteristics, within the epidemiological paradox.

Implications
This review points to several important considerations for interventions aimed at reducing sexual risk behaviors among adolescents as well as informs policymakers and health care providers. These findings imply that behavioral public health interventions to prevent engagement in sexual risk behaviors among both U.S.-born and foreign-born Latina/os may need to attend to multiple social-ecological processes, including family and peers, as well as cultural factors uniquely common within this community. Specifically, the integration of parent and family roles in interventions and an emphasis on their involvement in adolescents' development and maintenance of cultural values may decrease differences in sexual risk behaviors across acculturation, particularly those at increased risk from high acculturation backgrounds. Such integration of typical cross-cultural parenting practices with Latino culturally specific values may also help adolescents develop strategies to resist peer pressure and negative neighborhood influences to which Latino families are often exposed.
Moreover, awareness of the role of acculturation and sociocultural mechanisms within Latino health and health behaviors may provide a culturally sensitive guideline for health care providers. For example, acknowledging that differences in parenting practices and cultural values differ based on acculturation, as indicated by language use, nativity, or generational status, can enhance patient-provider interpersonal interactions and better address cultural-specific beliefs that may influence health outcomes. A clearer understanding of the epidemiological paradox and the underlying sociocultural mechanisms may also provide awareness among policymakers. Policies focused on sex education in public schools may need to address cultural values prominent in different groups.

Conclusions
This systematic review is the first, to our knowledge, that has examined the social and cultural underlying mechanisms of the epidemiological paradox in sexual risk behaviors among Latina/o youth. This systematic review, covering research published over 20 years (January 2000-October 2022), examined social and cultural mechanisms that were moderators and/or mediators of the epidemiological paradox of Latina/o adolescents' sexual risk behaviors. The Latino community comprises complex social and cultural factors at the individual, family, community, and societal levels that are critical when addressing adolescent sexual health needs. Latina/o adolescent sexual health is a growing field of research with multiple challenges that need to be addressed. This emerging research field provides contradictory evidence that does not yet clarify the explanatory mechanisms of the epidemiological paradox. Yet, the growing evidence suggests that social and cultural factors that vary by acculturation levels are nuanced predictors of sexual behaviors among Latina/o youth. As more evidence is accumulated, such findings may provide culturally sensitive and relevant intervention guidelines for reducing the high rates of sexual risk behaviors among this population.