Sexual Functioning in Adolescents with Major Depressive Disorder: A Prospective Study

Abstract

Objectives:

To examine the association between sexual functioning, depression and anxiety severity, and selective serotonin reuptake inhibitor (SSRI) use in adolescents.

Methods:

From September 2010 to December 2014, 15 to 20 year-old participants, either unmedicated or within a month of beginning SSRI treatment, completed the Beck Depression Inventory-II (BDI-II), Beck Anxiety Inventory (BAI), and the Changes in Sexual Functioning Questionnaire (CSFQ) at baseline and every four months, for up to two years. The DSM-IV-TR was used to determine presence of psychiatric disorders. Data regarding use of medications and hormonal contraception was collected. Polymorphisms of the 5HT2A and ABCB1 genes were genotyped. Linear mixed-effects regression models examined the association between depression and anxiety symptom severity, SSRI use, and sexual functioning, accounting for relevant covariates.

Results:

Two hundred sixty-nine participants (59% female, mean age 18.9±1.6 years, 71% with major depressive disorder) contributed to this analysis. After adjusting for age, sex, and duration in the study, depression severity, but not anxiety severity, was associated with lower CSFQ total scores (β= −0.13, p<.0001) and lower arousal, orgasm, and pleasure subscale scores (all β= −0.03, p<0.003). Higher SSRI doses were associated with lower orgasm subscale scores (β= −0.30, p<0.03). Hormonal contraceptive use was associated with higher CSFQ total scores (β= 0.97, p<0.003) and higher arousal (β= 0.25, p<0.009), desire (β= 0.24, p<0.001), orgasm (β= 0.27, p<0.02), and pleasure (β= 0.15, p<0.004) subscale scores. No significant genetic moderating effect was found.

Conclusions:

In adolescents, depression is associated with lower sexual functioning while SSRI use impairs orgasm.

INTRODUCTION

Sexual functioning is an important facet of life, with difficulties or dysfunction causing distress and reducing quality of life. However, little is known about sexual functioning in adolescents and emerging adults¹ despite this being a major developmental transition period.

As teens explore their sexual life, they are learning about what works and what does not. In fact, in 15 to 24 year-old French adolescents, 48% of females and 23% of males reported having impairment in at least one area of sexual functioning.² Similarly, in a recent two-year prospective study of 16 to 21 year-olds,³ 47% of female and 41% of male adolescents reported having a distressing sexual problem. Thus, understanding the extent and nature of sexual dysfunction in this age group is critical, given that sexual activity and functioning are continuing to develop and mature. Notably, the implications extend beyond sexual satisfaction as positive sexual self-concept has been associated with improved physical and psychological well-being.⁴

Even less is known about sexual functioning in adolescents with major depressive disorder (MDD). Although impaired sexual functioning is not a defining symptom of depressive episodes, it is quite prevalent, affecting more than 40% of adult women and 30% of men with MDD.⁵ Even in nonclinical samples, depression severity and sexual functioning are correlated, with desire being more affected than arousal or orgasm.^6,7 In the only study⁸ that, to our knowledge, examined sexual functioning in adolescents and young adults, the presence of a major depressive episode (MDE) was associated with lower sexual functioning overall as well as in the facets of desire, arousal, and orgasm, more specifically. Furthermore, males exhibited increasing impairment with worsening depression severity.

Importantly, the association between MDD and sexual functioning is complex, potentially confounded by antidepressant use. A recent review of the literature⁹ revealed that 35% of patients treated with SSRIs endorsed orgasmic dysfunction and 10% endorsed a sexual arousal disorder compared to 10% and 6%, respectively, reported in both the bupropion and placebo groups. However, while sexual adverse drug reactions (ADRs) to antidepressants have been extensively studied in adults,^6,10 only two studies examined this concern in adolescents. The first study¹¹ reported no cases of sexual dysfunction among 2,264 adolescents treated with SSRIs. Notably, this analysis used medical and pharmacy claims, well known to be constrained by under-reporting of ADRs.¹² Moreover, the rates of sexual dysfunction are substantially lower in studies relying on spontaneous reporting compared to those that formally assess for it.¹³ The second study⁸ consisted of a cross-sectional examination and also found no association following exposure to SSRIs lasting less than one month.

Moreover, studies in adults have investigated whether variants in genes related to antidepressant pharmacokinetics and pharmacodynamics moderate the emergence of sexual ADRs. In particular, there has been interest in the ATP-binding cassette, subfamily B, member 1 gene (ABCB1), which encodes for the drug efflux transporter p-glycoprotein (Pgp). Pgp is an efflux transporter expressed at the blood-brain barrier which functions to remove substrates, impacting their bioavailability in the brain. Substrates include commonly prescribed antidepressants such as citalopram, paroxetine, and sertraline.¹⁴ Genetic variants in the ABCB1 gene have been found to predict antidepressant response in patients receiving “Pgp substrate” medications.¹⁵ Relatively common polymorphisms exist that may impact the substrate specificity of the Pgp protein.¹⁶ For example, Bly et al.¹⁷ found that women with the rs1128503 TT genotype of the ABCB1 gene who received a Pgp-substrate antidepressant reported significantly lower sexual functioning as captured by the Changes in Sexual Functioning Questionnaire (CSFQ), than did those with the CT or CC genotypes. Additionally, given that serotonin (5HT) contributes to sexual arousal in females and modulates sexual functioning in males,^18,19 pharmacogenetic studies have also examined variants in genes involved in 5HT signaling. For instance, Bishop et al.¹⁸ found that adults treated with SSRIs with the GG genotype of the −1438 G/A variants of the 5HT2A receptor gene were significantly more likely to report impaired sexual arousal than their GA and AA genotype counterparts. To our knowledge, however, no such studies have been conducted in adolescents. Understanding the role genetic variants may play in the emergence of SSRI-induced sexual ADRs can impact treatment planning, particularly as treatment-induced sexual dysfunction affects quality of life,²⁰ potentially impeding treatment adherence.²¹

Thus, the purpose of this analysis was to examine sexual functioning in older adolescents with MDD over a two-year period as an extension to our previous cross-sectional study, and investigate the possible moderating effect of functional ABCB1 variants [C3435T (rs1045642), C1236T (rs1128503), and G2677A/T (rs2032582)] and 5HT2A variants (rs6311) on SSRI-induced sexual dysfunction. We hypothesized that depressive symptom severity would be associated with worse sexual functioning and that individuals with a greater number of T alleles in the ABCB1 gene and carriers of the GG genotype of the 5HT2A gene would report increased SSRI-induced sexual ADRs.

METHODS

Participants

Between September 2010 and December 2014, 15 to 20 year-old participants were enrolled in an observational study^22,23 examining the skeletal effects of SSRIs. Participants were either unmedicated or within a month of beginning SSRI treatment. Use of other psychotropics led to exclusion, with the exception of benzodiazepines (n=6 at intake), trazodone (n=4 at intake), α2-agonists (n=1 at intake), and psychostimulants (n=3 at intake). Exclusionary criteria included substance dependence, eating disorders, pregnancy, significant medical or surgical history, chronic use of medications potentially affecting bone metabolism (e.g., corticosteroids), and plans to move out of state in the following year. The study was approved by the local Institutional Review Board. Adult participants and the parent/guardian of minor participants provided written informed consent to the study, and the minor participants provided written assent.

Procedures

Participants completed a baseline visit, at which time demographic data were collected, and returned for follow-up visits every four months, for up to two years. They were contacted by phone monthly between in-person visits. At each encounter, data regarding the use of medications, including psychotropics and hormonal contraception (for females) as well as alcohol usage were collected. Adherence was based on self-report and pharmacy records. SSRI doses were converted into a common unit, whereby one unit was equivalent to a daily dose of 20 mg of fluoxetine, paroxetine, or citalopram, 50 mg of sertraline, or 10 mg of escitalopram.^22,23 At every in-person visit, participants also completed the Beck Depression Inventory (BDI-II),²⁴ and the Beck Anxiety Inventory (BAI).²⁵

Psychiatric diagnoses, following the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision,²⁶ incorporated information from a review of the medical records, the symptom rating scales, the National Institute of Mental Health (NIMH) Diagnostic Interview Schedule for Children IV (DISC-IV),²⁷ and an unstructured interview by a child psychiatrist. Sexual functioning was assessed using the 14-item CSFQ.²⁸ This self-completed questionnaire uses a 5-point Likert scale to measure the 5 dimensions of sexual functioning (i.e., pleasure, desire-frequency, desire-interest, arousal, and orgasm), with higher scores corresponding to better sexual functioning.²⁸

Genomic DNA from white blood cells (98.4%) or saliva (0.8%) [0.8% missing information regarding source] was extracted using Puregene (Gentra Systems, Minneapolis, Minnesota) or Oragene Kit (DNA Genotak, Kanata, Ontario, Canada). After DNA sample yield and purity were established through spectrophotometry and polymerase chain reaction (PCR) amplification, DNA samples were diluted to 20 ng/μL and stored at −20°C. PCR primers and determination of ABCB1 variants [C3435T (rs1045642), C1236T (rs1128503), and G2677A/T (rs2032582)] and 5HT2A variant (rs6311) genotypes followed. Established PCR products were visualized by electrophoresis on 1.8% agarose gels stained with ethidium bromide. PCR products were then sequenced using a Pyrosequencer (PyromarkMD) using standard protocols.

DATA ANALYSIS

The BDI-II total score excluded the sexual interest item in order to avoid inflating the association with sexual functioning. Because depression and anxiety are often comorbid and because little information is available about the potential impact of anxiety on sexual functioning, we examined the independent effects of the BDI-II and BAI scores. Given the repeated, albeit variable, number of observations per study participant, linear mixed-effects regression models were fitted to examine the associations between depressive and anxiety symptom severity and SSRI dose, on the one hand, and sexual functioning, on the other.²⁹ All models included adjustment for age (years) at study entry, duration of time in the study,³⁰ and sex. Participant-specific random intercepts and slopes were modeled assuming an unstructured covariance matrix. Maximum likelihood (ML) methods were used for estimation, which yields unbiased estimates under the assumption that the missing data mechanism is ignorable.³¹ Model assumptions of linearity, homogeneity of variance of the residuals, and normality of the residuals and random effects were checked using appropriate residual plots.²⁹

Given that hormonal contraceptive use is associated with both depression³² and sexual functioning,³³ we repeated the analyses accounting for use of hormonal contraception in females. In addition, the moderating effect of genetic variants was examined, using the same overall models. SSRIs were grouped based on existing literature regarding their potential activity as Pgp substrates in vivo.¹⁴

Previous research³⁴ has shown that the ABCB1 variants are in linkage disequilibrium and that presence of a T allele in one or more of the three variants genotyped is associated with reduced Pgp expression, which may result in increased drug toxicity. Furthermore, previous research³⁵ has also shown the presence of at least two T alleles to be associated with differences in medication response for Pgp substrates. Therefore, participants were divided based on the presence of < or ≥ 2 T alleles. In order to optimize statistical power, sertraline, citalopram, and escitalopram were combined in a single group as all three are Pgp substrates.¹⁵ Subsequently, six haplotype/SSRI groups were created based on the combination of ABCB1 haplotype groups and SSRI groups. In order to examine the potential moderating effect of ABCB1 haplotype on sexual ADRs, a haplotype/SSRI group by SSRI dose interaction term was added to the model. In order to minimize type 1 error, the moderating effect of the genetic variants was examined only in models where SSRI dose was significantly associated with sexual functioning.

All analyses were conducted using SAS version 9.4 (SAS Institute Inc., Cary, North Carolina) with a significance level of P ≤ 0.05.

RESULTS

Participant Characteristics

Of 279 participants enrolled in the study, 263 contributed to this analysis, after excluding those with a psychotic disorder, bipolar disorder, autism spectrum disorder, a genetic condition, or those with invalid CSFQ data. Table 1 summarizes the demographic and clinical characteristics of the participants.

Table 1:

Baseline Demographic and Clinical Characteristics of the Participants

	Participants (n=263)
	Mean or N	Stda or %
Age, years	18.9	1.6
Female Sex	156	59
Race
Caucasian	233	89
African American	12	5
Ethnicity
Hispanic	22	8
Non-Hispanic	241	92
Hormonal Contraceptionb	77	49
Progesterone	2	1
Combination	54	35
Other	21	13
SSRI
Citalopram/Escitalopram	50	19
Fluoxetine	37	14
Sertraline	32	12
Paroxetine	4	2
Baseline SSRI Dose, unit/dayc	0.4	0.6
Major Depressive Disorder	185	71
Generalized Anxiety Disorder	79	30
Social Anxiety Disorder	72	27
Panic Disorder	18	7
BAI	8.5	8.7
BDI	11.2	10.5
CSFQ
Total Score	45.9	11.2
Desire Sub-scale	6.0	2.0
Interest Sub-scale	8.0	2.6
Arousal Sub-scale	10.0	3.0
Orgasm Sub-scale	9.6	3.8
Pleasure Sub-scale	3.1	1.3
ABCB1 Haplotypes
No T	54	21
1T	37	14
2T	16	6
>2T	156	59
HT2RA Variantsd
GG	197	84
AG	37	16
AA	0	0

^a:Standard deviation

^b:Analysis restricted to female participants.

^c:One SSRI unit was equivalent to 10mg of escitalopram, 20mg of fluoxetine, citalopram or paroxetine, and 50mg of sertraline.

^d:Genotyping of this variant of the HTR2A gene failed in 29 cases.

Abbreviations:

BAI, Beck Anxiety Inventory;

BDI, Beck Depression Inventory

CSFQ, Changes in Sexual Functioning Questionnaire

SSRI, selective serotonin re-uptake inhibitor

ABCB1 Haplotypes: Haplotypes created based on the number of T alleles in 3 variants of The ATP-binding cassette, subfamily B, member 1 gene.

Sexual Functioning, Depression, Anxiety, and SSRI Dose

After adjusting for age (β= 0.1297, p≤0.0003) , sex (β= 0.3220, p<0.006), and time in the study (β= 0.1456, p<0.005), higher BDI-II scores were associated with lower CSFQ total scores (β= −0.13, p<.0001) as well as lower scores on the arousal (β= −0.03 , p<0.003), orgasm (β= −0.03, p<0.002), and pleasure (β= −0.03, p<.0001) subscales (Table 2, Model 1). In contrast, higher SSRI doses were only associated with reduced orgasm subscale scores (β= −0.30, p<0.03) [Table 2, Model 1].

Table 2:

Parameter Estimates (95% Confidence Limits) for BDI-II, BAI, and SSRI dose from Linear Mixed-Effects Regression Models Predicting Sexual Functioning^a

		Total CSFQ Score	Arousal Sub-scale	Desire Sub-scale	Interest Sub-scale	Orgasm Sub-scale	Pleasure Sub-scale
Model 1	BDI-II	-0.13 (−0.19, −0.07)	-0.03 (−0.05, −0.01)	−0.01 (−0.02, 0.01)	−0.01 (−0.03, 0.00)	-0.03 (−0.05, −0.01)	-0.03 (−0.04, −0.02)
	SSRI	−0.22 (−0.95, 0.51)	0.06 (−0.16, 0.28)	0.03 (−0.13, 0.20)	0.06 (−0.15, 0.26)	-0.30 (−0.54, −0.05)	−0.01 (−0.12, 0.11)
Model 2	BAI	-0.12 (−0.18, −0.05)	-0.02 (−0.04, −0.00)	−0.01 (−0.02, 0.01)	−0.01 (−0.03, 0.00)	-0.03 (−0.05, −0.00)	-0.02 (−0.03, −0.01)
	SSRI	−0.30 (−1.03, 0.43)	0.03 (−0.19, 0.26)	0.03 (−0.14, 0.19)	0.05 (−0.15, 0.26)	-0.32 (−0.56, −0.07)	−0.04 (−0.16, 0.08)
Model 3	BDI-II	-0.11 (−0.18, −0.04)	-0.03 (−0.05, −0.00)	−0.01 (−0.02, 0.01)	−0.01 (−0.03, 0.01)	-0.03 (−0.05, −0.00)	-0.03 (−0.05, −0.02)
	BAI	−0.04 (−0.12, 0.04)	−0.00 (−0.03, 0.02)	−0.00 (−0.02, 0.02)	−0.01 (−0.03, 0.01)	−0.01 (−0.04, 0.02)	0.01 (−0.01, 0.02)
	SSRI	−0.20 (−0.93, 0.53)	0.06 (−0.16, 0.28)	0.03 (−0.13, 0.20)	0.06 (−0.15, 0.27)	-0.29 (−0.54, −0.04)	−0.01 (−0.13, 0.11)
Model 4	BDI-II	-0.11 (−0.21, −0.02)	-0.03 (−0.05, −0.00)	−0.01 (−0.03, 0.01)	−0.00 (−0.03, 0.02)	-0.03 (−0.06, 0.00)	-0.03 (−0.05, −0.01)
	BAI	−0.02 (−0.13, 0.08)	0.00 (−0.03, 0.03)	0.01 (−0.02, 0.03)	−0.00 (−0.03, 0.02)	−0.01 (−0.04, 0.03)	0.01 (−0.01, 0.03)
	SSRI	−0.22 (−1.14, 0.70)	0.05 (−0.22, 0.32)	−0.08 (−0.28, 0.12)	0.11 (−0.13, 0.35)	−0.26 (−0.58, 0.05)	0.00 (−0.14, 0.15)
	Contraceptive	0.97 (0.34, 1.60)	0.25 (0.06, 0.44)	0.24 (0.10, 0.38)	0.09 (−0.08, 0.25)	0.27 (0.05, 0.49)	0.15 (0.05, 0.26)

^a:The base model included age, sex, and time in the study.

Abbreviations:

BAI, Beck Anxiety Inventory

BDI-II, Beck Depression Inventory II

CSFQ, Changes in Sexual Functioning Questionnaire

SSRI, selective serotonin re-uptake inhibitor

Comparable findings were observed with the BAI score, albeit the parameter estimates were smaller than with the BDI-II (Table 2, Model 2). When both the BDI-II and BAI scores, along with the same other covariates, were concurrently entered into the models, only the BDI-II score remained significantly associated with sexual functioning (β= −0.11, p<0.003) [Table 2, Model 3]. Notably, accounting for SSRI adherence did not alter the findings and linear mixed-effects model assumptions were satisfied (results available from the authors upon request).

When individual SSRI agents were examined, potentially significant differences emerged only with regard to the pleasure (p<0.03) and desire subscales (p=0.08). In both instances, the use of citalopram/escitalopram was associated with lower scores compared to sertraline (β= −0.44, p<0.02, for pleasure and β= −0.53, p<0.04, for desire, respectively) and fluoxetine (β= −0.53, p=0.005, for pleasure and β= −0.53, p<0.05, for desire, respectively).

Sexual Functioning, Depression, and Hormonal Contraceptive Use

When the analyses were restricted to females and hormonal contraceptive use was accounted for, the association of depression severity with sexual functioning was not altered (β= −0.13, p<0.001). However, contraceptive use was independently associated with higher total scores on the CSFQ (β= 0.97, p<0.003) as well as on the arousal (β= 0.25, p<0.009), desire (β= 0.24, p<0.001), orgasm (β= 0.27, p<0.02), and pleasure (β= 0.15, p<0.004) subscales (Table 2, Model 4).

Moderating Effect of Genetic Variants on Sexual Functioning

Genotype and allele distributions for the 5HT2A −1438 G/A SNP are presented in Table 1. The allele frequencies for the candidate gene loci did not deviate from Hardy-Weinberg equilibrium (X²=1.72, 1df, p=0.19).

After adjusting for age, sex, time in the study, BDI-II score, and SSRI dose, neither the main effect of the 5HT2A variants nor their interaction effect with SSRI dose had significant association with the orgasm subscale of the CSFQ (both p>0.10).

Table 3 shows the number of participants per haplotype groups (i.e., < 2T vs. ≥ 2T) in each medication group at baseline. After adjusting for the same covariates, while higher SSRI doses remained related to lower orgasm subscale scores of the CSFQ (β= −1.31, 95% confidence limits: −2.33, −0.29), neither the main effect of the ABCB1 haplotype/SSRI grouping nor its interaction effect with SSRI dose were significantly associated with the orgasm subscale (both p >0.20). These results remained unchanged after accounting for adherence and after restricting the sample to non-Hispanic whites (p >0.30).

Table 3.

Number (%) of Participants by the ATP-binding cassette, subfamily B, member 1 gene Haplotype and Medication, at Baseline

	< 2T	≥ 2 T
Sertraline/Citalopram/Escitalopram	31 (12)	59 (22)
Fluoxetine	14 (5)	22 (8)
No Selective Serotonin Reuptake Inhibitor	46 (17)	91 (35)

DISCUSSION

This study expands on our previous work⁸ examining the association between depression, SSRI use, and sexual functioning by using a prospective design. Overall, we confirmed that depressive symptom severity, but not that of anxiety, was associated with lower sexual functioning. In contrast to our earlier cross-sectional data, we found that SSRI use was associated with specific impairment in orgasm. Finally, we found no consistent moderating effects of 5HT2A genetic variants or ABCB1 haplotypes on sexual functioning in participants treated with SSRIs.

In adults, more severe depression is associated with decreased sexual functioning. To our knowledge, this has not been prospectively examined in adolescents. Importantly, we found that depression was associated with lower functioning in the arousal, orgasm, and pleasure aspects of the sexual cycle. This contrasts our cross-sectional work which showed a relationship with impairment in desire, but not pleasure. Several factors may account for this discrepancy including methodological differences or the way the CSFQ captures these two facets – pleasure is measured using a single item whereas desire is measured using five, including frequency and interest.²⁸ The items related to frequency of sexual activity do not take into account other possible causes of less frequent sexual activity including pain during intercourse, which has also been shown to hinder sexual functioning.² Some studies³⁶ indicate that for late adolescents, sexual pleasure is a large motivator for engaging in sexual activity, including “hook-ups,” which may not be optimally captured with a single questionnaire item.

Similar to findings in adults,¹⁰ this investigation found SSRI use was associated with impairment in orgasm in our adolescent participants which was not surprising given their successful use in treating premature ejaculation³⁷ due to serotonin’s physiological role in mediating orgasm.^19,38 Of note, the association between SSRI use and lower orgasm subscale scores was not seen in the female-restricted analysis most likely due to the smaller sample size, given that the slope estimates were only minimally different (Table 2, Model 4). In addition, preliminary analyses suggest that differences between citalopram and escitalopram, on the one hand, and sertraline and fluoxetine, on the other, may exist with the two former medications more adversely affecting the pleasure and desire facets of sexual functioning. Whether these findings are due to differences in pharmacokinetics and pharmacodynamics is unclear, as this research is still evolving, partly due to the fact that identification of specific therapeutic ranges for this class of medication is unknown. This is especially true for sexual dysfunction, as the mechanism underlying potential differences between agents remains unknown.³⁹

Performance anxiety has been suggested as a possible cause for poor sexual functioning.^40,41 However, we found no independent relationship between anxiety symptom severity and sexual functioning, consistent with our earlier finding.⁸ Importantly, anxiety may have variable effects on sexual arousal, with potential facilitation through increased attention to sexual cues⁴¹ or via misinterpretation of anxious arousal as sexual arousal given the overlap of bodily sensations.⁴² One shortcoming of these studies, however, is their failure to account for comorbid depression, as we did.

Hormonal Contraception

Our results indicate that hormonal contraception use is associated with better CSFQ scores. This is consistent with the baseline findings in this cohort⁸ and overlaps with the results from Ott et al.,⁴³ who found a reduction in sexual interest in adolescents who discontinued OCPs. Of note, hormonal contraception use is associated with greater sexual problems and less frequent sexual activity in adult women.⁴⁴ While this difference may be related to biological variances, it may also be due to alleviation of anxiety about unwanted pregnancy in adolescents who are initiating sexual activity.⁴⁵ Further examination is warranted given evidence implicating hormonal contraceptive use by adolescents in the emergence of depression.³²

Genetics

Several recent reviews have examined the opportunities and challenges offered by pharmacogenomics in general and as they relate to child and adolescent psychiatry and SSRI use.^46–48 In contrast to some published reports,⁴⁹ we found no association between the 5HT2A (rs6311) variant and sexual functioning. This divergence may be due to differences in study design, including whether the participants were actively depressed when sexual functioning was assessed or had had sexual dysfunction prior to initiating antidepressant treatment.¹⁸ Moreover, racial and ethnic composition of the participants may play a role, given the variability in 5HT2A variants’ moderating effect on treatment response based on racial/ethnic backgrounds.⁵⁰ Similarly, we found no significant moderating effect of the ABCB1 haplotype.

Limitations

There are several limitations to this study. Firstly, the CSFQ has been validated in adults, but not adolescents. The CSFQ defines sexual activity broadly beyond intercourse, including masturbation and fantasizing, and some studies⁵¹ show that adolescents’ specific definition of sex can vary widely. Thus, collecting specific information regarding sexual activity would have allowed better interpretation of the results. In addition, the BDI-II and BAI may not be the optimal scales to capture, respectively, depression and anxiety symptom severity in youths.⁵² This may account for the lack of an independent association between anxiety severity and sexual functioning. Our study population consisted primarily of whites, which reflects the local population. However, the impact of MDD on sexual functioning should also be examined in more diverse populations, given differences in sex initiation across races/ethnicities.^53,54 In addition, adolescence is a time where individuals begin to explore not only their sexuality, but their identity. Our study does not take into account gender identity differences and the impact this may have on sexual exploration. Finally, larger studies are likely needed for well-powered pharmacogenetic analyses examining additional variants, particularly of the CYP2D6 and CYP2C19 genes, given their potential associations with other side effects in youths, often via a dose-dependent effect.⁵⁵ Future studies should also be adequately powered to explore differences between individual SSRIs.

In sum, in this longitudinal study, we confirm that depression severity is associated with impaired sexual functioning in both male and female adolescents. Moreover, the SSRI effect appears limited to orgasm. As adolescents mature, sexual functioning features more prominently in their lives, requiring clinicians to attend to this important facet of development. Given the propensity of adolescents to avoid spontaneously reporting on their sexual functioning,¹³ it is imperative that providers initiate discussions about this topic. Proper assessment may result in treatment implications, including avoiding treatment non-adherence by reducing the dose, switching medications, or even discontinuing psychotropics in favor of non-pharmacological interventions.⁵⁶

Clinical Points.

• The association between depression, anxiety, and sexual functioning is complex, potentially confounded by selective serotonin reuptake inhibitor (SSRI) use. There is limited research in adolescents and young adults despite this being a major developmental transition period.

• Otherwise medically healthy 15 to 20 year-old participants within a month of starting an SSRI or who were unmedicated were enrolled in a 2-year prospective study, during which symptom severity, sexual functioning, and SSRI use were monitored. Depression severity rather than anxiety severity was associated with worsening sexual functioning, while SSRI use was implicated in worsening orgasmic function.

• Clinicians should ask adolescents and young adults about their sexual functioning (e.g. desire, interest, arousal, and orgasm) at baseline and during depression treatment and address the concerns when they arise.