Associations of future cognitive decline with sexual satisfaction among married older adults

Abstract

Objectives:

This study sought to explore changes in longitudinal cognitive status in relation to baseline measures of intimacy and sexuality in cognitively intact, married older adults.

Methods:

Baseline intimacy and sexuality survey data from 155, cognitively intact, married, older adults were collected using a novel survey instrument that explored the domains of: 1) romance with one’s partner, 2) sexual satisfaction, 3) beliefs about sexuality, and 4) social support and emotional intimacy. These data were analyzed in relation to change in cognitive status over a 10-year follow-up period using binary logistic regression modeling. Exploratory factor analysis was used to assess the shared variance of survey items attributable to intimacy and sexuality without specification of an a priori hypothesis regarding the association of intimacy and sexuality with future change in cognitive status.

Results:

Over the 10-year study period, 33.5% (n=52) of individuals developed cognitive impairment. Participants with greater sexual satisfaction scores at baseline were statistically less likely to convert from cognitively intact to mild cognitive impairment or dementia in the future (p = 0.01). The domains of romance with one’s partner, beliefs about sexuality, and social support/emotional intimacy were not predictive of future longitudinal changes in cognitive status.

Conclusions:

Sexual satisfaction is associated with longitudinal cognitive outcomes in cognitively intact, married, older adults.

Clinical implications:

Clinicians should routinely assess for sexual satisfaction among older adults and refer to appropriate providers, such as couples or sex therapists, when appropriate.

Introduction

Older adults remain sexually active and intimate well into old age (Freak-Poli et al., 2017). Although research on sexuality and intimate relationships in older adulthood has increased in recent years, little is known about the relationship between intimacy, sexuality, and cognitive health over time. Several recent studies have reported positive associations between frequency of sexual activity and better cognitive health in older adults (Allen, 2018; Wright & Jenks, 2016; Wright, Jenks, & Demeyere, 2019). However, there is a lack of longitudinal research assessing this relationship. The breadth of the current literature has been stunted due to a narrow focus on sexual activity alone, excluding other pertinent psychosocial aspects of intimate romantic relationships (Araujo et al., 2004). Psychosocial aspects, such as physical and emotional closeness, positive emotions, and romance, in addition to sexuality, (Moss and Schwebel, 1993; Tolstedt and Stokes, 1983) are critical in defining intimacy.

It also remains unclear whether there is a causal relationship between sexuality/intimacy and cognitive health or vice versa, and to what extent other confounding factors—such as physical wellness and mental health—play in such associations. Sexuality and intimacy involve a nexus of biological and psychosocial factors, including health, well-being, quality of life, and physical activity endurance (Wright & Jenks, 2016). Each of these factors are known to be associated with cognitive aging. For instance, chronic illness is a strong predictor of cognitive decline (Tilvis et al., 2004) and is also associated with reduced sexual activity (Bancroft, 2007), sexual dysfunction (Lindau et al., 2007), and lower sexual satisfaction (Flynn et al., 2016). Thus, individuals who are more physically and cognitively healthy may also be engaging in more satisfying sexual activity.

In addition, research on intimacy in older adulthood has focused on the biological components of sexual functioning, failing to account for social, psychological, and relationship factors that are pertinent to an individual’s expression of sexuality and intimate relationships (DeLamater & Karraker, 2009). Evidence suggests that older adults who frequently engage in sexual activity, including kissing, petting, and fondling, have higher levels of well-being and life satisfaction than those who are not sexually active (Smith, et al., 2019). Accordingly, since greater well-being is linked to better cognitive function (Allerhand et al., 2014), sexual activity may positively impact cognitive health indirectly through its impact on well-being (Wright & Jenks, 2016). Conversely, lower levels of sexual activity and sexual satisfaction are associated with poorer overall well-being, increased likelihood of experiencing loneliness, depression, and other psychiatric disorders, all of which are associated with poorer cognitive functioning (Allerhand et al., 2014; Sachs-Ericsson et al., 2005; Tilvis et al., 2004).

The present study assesses the relationship of baseline sexuality and intimacy (as defined by the construct domains of: romance, sexual satisfaction, beliefs about sexuality, and social support/emotional intimacy) and change in cognitive status over a 10-year period in married older adults, considering and addressing many of the confounds inherent in the constructs of sexuality and intimacy highlighted in the existing literature. In essence, this study seeks to address the question of whether aspects of sexuality and or intimacy among older married adults are predictive of change in cognitive status over time, providing insights into a potential causal relationship that has not previously been explored.

Methods

Data for this retrospective analysis were collected as part of the University of Kentucky Alzheimer’s Disease Center (UKADC) longitudinal observational study of aging. Participants in the study consent to undergo medical, neuropsychological, and neurological examinations once per year. Details of this cohort and the assessment procedures have been previously published (Schmitt et al., 2012 In 2009, all active members of the UKADC cohort (N = 668) were invited to complete a survey assessing interpersonal relations, intimacy, and sexuality. The survey was sent via mail with a cover letter, and a stamped self-addressed envelope to facilitate return. All procedures were approved by the University of Kentucky IRB.

Participants

Inclusion criteria for the present study were (a) being cognitively unimpaired at the time of completing the intimacy survey (consensus diagnosis as normal), (b) completing at least two UKADC study visits between 2009 and 2020 (to detect change in cognitive status over time), and (c) being married at the time of survey completion. Individuals were considered cognitively unimpaired and excluded from the present study if they were given a consensus diagnosis of mild cognitive impairment (MCI) or dementia (see cognitive status section below for details of the research diagnostic process). While we recognize that sexuality and intimacy are important regardless of marital status, we lacked details regarding relationship status for respondents who were not married and so we excluded such participants in order to avoid drawing potentially inaccurate conclusions about intimacy in such participants.

Measures

Cognitive status

Respondents were categorized as cognitively intact or impaired (meeting criteria for mild cognitive impairment or dementia) by consensus diagnostic procedures described below. This process was repeated annually or until death for all study subjects. Following each UKADC study visit, cognitive status was determined at interdisciplinary consensus diagnostic meetings consisting of a neuropsychologist, examining clinician, and research associate/psychometrist that conducted the testing protocol. Consensus diagnosis for MCI followed the recommendations of the 2^nd International Working group on MCI (Winblad et al. 2004) and the diagnosis for dementia was based on the criteria set forth in the Diagnostic and Statistical Manual for Mental Disorders (DSM-4; American Psychiatric Association, 2000). The present study focuses on the broader assessment of cognitive impairment at either the MCI or dementia stage rather than exploring etiological diagnoses that may have provided limited data given the small number of participants with longitudinal change over the course of the present study.

Intimacy and sexuality

Development of the interpersonal relations, intimacy, and sexuality survey (IRIS) began at the UKADC in 2008. At the time, no measure available assessed a holistic view of intimacy, including beliefs about sexuality, sexual satisfaction, and physical and emotional intimacy with others. An expert panel selected items from existing instruments including the Friendship Scale (Hawthorne, 2006), the Intimacy Scale (Sinclair & Dowdy, 2005), the Social and Emotional Loneliness Scale (DiTommaso et al., 2004), and the Sexuality Scale (Snell & Papini, 1989). To pilot the IRIS, 25 consecutive consenting UKADC participants known to be cognitively unimpaired completed the questionnaire in person at their annual study visit. Debriefing interviews were used to obtain feedback and verbal probes were used to assess interpretation of items. Panel members revised items based on feedback from participants.

The final iteration of IRIS had four subscales. The romance subscale assesses physical and emotional intimacy with one’s partner. Sexual satisfaction assesses perception of the quality of their sex life. The beliefs about sexuality subscale assesses personal attitudes toward sex and their own sexuality. Social support/emotional intimacy subscale assesses emotional support from individuals in one’s social network as well as engagement in social activities. During piloting and data collection, respondents rated their level of agreement with each item stem on a Likert scale of 0 to 4, where 0 = strongly disagree. Some items were reverse coded. Scores for each domain were then computed by averaging all items responded to from that domain.

The IRIS was completed by 322 older adults in early summer 2009; 74 were cognitively impaired and 248 were cognitively healthy. The IRIS was evaluated using data from all 248 cognitively unimpaired respondents. Exploratory factor analysis (EFA), with a four-factor solution and a promax oblique rotation (assuming factors were interrelated) found that 21 of 29 items (72.4%) had the highest loadings for the subscale they were grouped with by the panel. The items were re-grouped accordingly using a threshold for loading of 0.30, with four items having no factor loading > 0.30 (sex outside of marriage, feeling guilty about sex, STI concerns, and sex activity is duty of partner). The four-factor solution explained 42.5% of IRIS variance and internal consistency of the revised factors was good (variance explained, Chronbach’s α): romance (14.3%, 0.88), sexual satisfaction (12.4%, 0.83), beliefs about sexuality (8.5%, 0.71), and social support (7.3%, 0.73).

Covariates

Covariates in the analysis included age (years), sex (male or female), education (years), cognitive composite scores, objective health, and subjective health.

Cognitive composite scores

A global cognitive composite score was used in the analysis to account for general cognitive functioning at the time of completing the IRIS. Including the global composite in the analysis allows for the unique association of intimacy and future cognitive decline. At each UKADC visit the Uniform Data Set (UDS) neuropsychological test battery (Weintraub et al., 2009) was administered. For the present study, raw test scores from this visit most-proximal to IRIS completion were transformed to z-scores adjusted for age, education, and sex according to established standards (Shirk et al., 2011). The global cognitive composite score was computed as the average of the z-scores of all the UDS tests, where higher values indicated greater cognitive functioning.

Objective health

The Functional Comorbidity Index (FCI; Groll et al., 2005) was used as a measure of objective health. The FCI is a validated index of 18 items that has been shown to accurately classify physical functioning. The FCI was computed as the number of the following conditions recorded in the medical record: asthma, COPD, angina, congestive heart failure, myocardial infarction, neurological conditions (Parkinsonism, seizures, or brain tumor), stroke or TIA, peripheral vascular disease, diabetes, ulcer or acid reflux, depressive episode within 2 years, anxiety or panic disorder, vision impairment, hearing impairment, back problems (spinal stenosis or any back surgery), and obesity (BMI > 30). Arthritis and osteoporosis were not available for this time period in the UK-ADC database and were not used in computation of the FCI. FCI is an integer with greater values indicating greater medical comorbidities.

Subjective health

Subjective health was measured by the eight subscales of the SF-36 (Ware & Sherbourne, 1992): physical functioning, role limitations due to physical health, bodily pain, general health, vitality, social functioning, role limitations for emotional health, mental health. The SF-36 is a validated instrument with 36 items widely used as an indicator of patient outcomes. Raw item responses were recoded per the published guidelines and the mean of recoded items was taken for each subscale score (Hays et al., 1995). Each subscale was continuous, with a range of 0 to 100, where greater values reflect greater subjective health. Summary scores for mental and physical health components were computed by the standardization and aggregation process outlined by Ware (1993).

Analysis

T tests and chi-square tests were conducted to assess for differences between continuous and categorical demographic and health variables, respectively. Binary logistic regression models were conducted to assess whether the four aspects of intimacy were predictive of change in cognitive status over 11 years. Significance was set at 0.05 in this exploratory analysis. All analyses were conducted using SPSS Statistical Software.

Results

Of the 668 participants who were mailed surveys, 310 (46.4%) responded. There were no significant differences between respondents and non-respondents in terms of age or sex, but non-respondents had lower levels of education than respondents t(666) = −2.15, p = .023 and a higher percentage of non-respondents had a diagnosis of cognitive impairment compared to respondents, χ² (1, N = 668) = 91.32, p < .001. Of the 310 individuals who returned surveys, 155 individuals met the inclusion criteria. Reasons for exclusion included baseline diagnosis other than cognitively unimpaired (n = 70) and being unmarried (n = 85). Respondents who met inclusion criteria (N = 155) were between the ages of 55 and 95 (M = 75.59, SD = 6.58), predominantly White (98.7%), highly educated (74.8% obtained either a bachelor’s or postgraduate degree), and evenly distributed between sexes (54.2% female). Data on gender identity and sexual orientation of respondents were not available.

Participants’ health status

Objective and subjective health was similar between participants who remained cognitively stable and those who cognitively declined. Specifically, there were no significant differences in FCI scores between participants who remained cognitively intact (M = 1.22, SD = 1.43) and those who transitioned to consensus diagnoses of MCI or dementia over the follow up period (M = 1.34, SD = 1.13), t(152) = −0.54, p = .170. Furthermore, there were no statistical differences in mental health subscale scores on the FCI and SF-36 between participants who cognitively declined (M = 53.02, SD = 5.95) versus those who remained cognitively stable (M = 52.91, SD = 6.16), t(143) = −0.10, p = .963. Likewise, there were no statistical differences in physical health subscale scores on the FCI and SF-36 between participants who cognitively declined (M = 45.55, SD = 8.53) versus those who remained cognitively stable (M = 46.09, SD = 9.68), t(143) = 0.34, p = .371. Mean scores for the FCI and SF-36 mental and physical health subscale scores are displayed in Table 2.

Table 2.

Descriptive characteristics of participants in relation to follow-up cognitive status (N = 155)

	Overall (N = 155)		Cognitively Stable (n = 102)		Cognitively Declined (n = 53)
Characteristic	n	%	n	%	n	%
Race
Black or African American	2	1.3	2	1.9	0	0.0
White	153	98.7	100	98.1	53	100.00
Sex
Female	84	54.2	58	56.9	26	49.1
Male	71	45.8	44	43.1	27	50.9
Widow Status	32	20.6
	M	SD	M	SD	M	SD
Age (years)	75.59	6.58	75.16	7.10	76.48	5.53
Education (years)	16.65	2.67	17.02	2.46	15.94	2.94
Cognitive composite	0.07	0.77	0.07	0.78	0.07	0.77
Functional Comorbidity Index (FCI)	1.26	1.33	1.22	1.43	1.34	1.13
SF-36 Mental Health	52.95	6.07	52.91	6.16	53.01	5.95
SF-36 Physical Health	45.91	9.28	46.09	9.68	45.54	8.53

Longitudinal change in cognitive status

Fifty-two participants (33.5%) who were cognitively intact at baseline received a diagnosis of MCI between 2010 and 2020. Participants transitioned to a diagnosis of MCI between 1 and 10 years after the administration of the intimacy survey (M = 4.62, SD = 3.02). Of these, 20 participants (38.5%) progressed to dementia, between 3- and 9-years post-baseline (M = 5.52, SD = 2.11). One participant who was cognitively intact at baseline did not receive a diagnosis of MCI at any point, but instead transitioned to dementia 4 years after the administration of the survey. Descriptive characteristics of participants including cognitive composite scores are presented in Table 3.

Table 3.

Mean scores for Intimacy Domains, by group (N = 155).

	Overall (N = 155)		Cognitively Stable (n = 102)		Cognitively Declined (n = 53)
Domain	M	SD	M	SD	M	SD	d	95% CI
Beliefs about Sexuality	2.36	0.75	2.44	0.77	2.20	0.69	0.343	[ 0.00, 0.68]
Romance	2.96	0.77	3.06	0.66	2.76	0.92	0.408*	[ 0.07, 0.75]
Sexual Satisfaction	2.31	0.79	2.47	0.73	2.00	0.81	0.624***	[ 0.28, 0.97]
Social Support	3.23	0.55	3.24	0.55	3.20	0.54	0.045	[−0.29, 0.38]

^*p < .05.

^**p < .01.

^***p < .001.

Although all participants were married at baseline, 20.6% became widowed at some point over the 11-year time period. A higher percentage of participants who experienced a spouse’s death were diagnosed with MCI or dementia (46.9%) compared to those who did not experience a spouse’s death (30.9%), however this difference was not statistically significant, χ² (1, N = 155) = 2.88, p = .090. Those who cognitively declined had significantly lower levels of education than those who remained cognitively stable, t(153) = 2.14, p = .017. Age and sex were unrelated to future cognitive status.

Relationship between IRIS domains and cognitive status

Relationships between baseline IRIS scores and change in cognitive status depended on domain. Participants who remained cognitively intact at all follow-up visits reported higher IRIS romance scores than those who subsequently experienced a change in cognitive status, t(101) = 2.36, p = .020, d = .408. Similarly, participants who remained cognitively intact over the follow up period reported higher IRIS sexual satisfaction scores than participants who subsequently experienced a change in cognitive status, t(100) = 3.66, p < .001, d = 0.624. The IRIS domains beliefs about sexuality and social support/emotional intimacy were similar between participants that remained cognitively intact and those that subsequently experienced a change in cognitive status. Mean scores for each domain, including effect sizes, are displayed for both groups in Table 3.

Binary logistic regression models (see Table 4) were used to assess the association between intimacy scores and likelihood of future change in cognitive status, Nagelkerke R² = .21, χ² (8, N = 152) = 25.19, p = .001. Notably, after controlling for age, sex, education, loss of a spouse, baseline cognitive composite score, objective health (FCI), and subjective health (SF-36), each unit increase in IRIS sexual satisfaction score significantly reduced the odds of a future MCI or dementia diagnosis (OR = 0.39, CI [0.19, 0.81]). Romance, beliefs about sexuality, and social support/emotional intimacy IRIS domain scores were not predictive of future change in cognitive status after controlling for the aforementioned variables. Taken together, these data indicate that among married older adults, lower sexual satisfaction scores on the IRIS are related to an increased likelihood of future change in cognitive status to MCI or dementia.

Table 4.

Binary logistic regression predicting change in cognitive status (N = 142, declined = 33.5%).

Predictor	B	SE	p	OR	95% CI
Factors
Beliefs about sexuality	0.18	0.33	.584	1.20	[0.63, 2.29]
Romance	−0.48	0.36	.186	0.62	[0.31, 1.26]
Sexual satisfaction	−0.95	0.37	.011	0.39	[0.19, 0.81]
Social support	0.95	0.50	.059	2.59	[0.97, 6.93]
Respondent Characteristics
Age	0.06	0.37	.110	1.06	[0.99, 1.14]
Cognitive composite	−0.26	0.27	.335	0.77	[0.46, 1.31]
Education	−0.19	0.10	.046	0.83	[0.69, 1.00]
FCI (objective health)	−0.15	0.18	.404	0.86	[0.61, 1.22]
Sex	−0.76	0.50	.127	0.47	[0.18, 1.24]
SF-36 mental component	−0.02	0.04	.554	0.98	[0.91, 1.05]
SF-36 physical component	0.00	0.03	.906	1.00	[0.95, 1.06]
Widow status	0.51	0.53	.333	0.33	[0.59, 4.69]

Note. CI = confidence interval for odds ratio (OR).

Discussion

Increasing evidence supports the importance of sexuality and intimacy in maintaining optimal health and cognitive function in the aging population. The present study investigated the relationship between various aspects of sexuality and intimacy in relation to longitudinal cognitive status among a cohort of 155 married older adults. Our results indicate that respondents with higher sexual satisfaction scores were significantly less likely to be diagnosed with MCI or dementia over a 10-year follow up period. This finding was significant after controlling for subjective and objective health, indicating a direct link between sexual satisfaction and cognitive health outcomes. Although there was no direct measure of frequency of sexual activity, the results of this study highlight the importance of accounting for the psychosocial aspects of sexual relationships in relation to cognitive health rather than overt sexual activity in and of itself. These data suggest that routine assessment of sexual satisfaction, and where appropriate, referral to health care specialists such as couples or sex therapists may prove beneficial in maintaining cognitive health for the aging population.

As there can be many aging-related confounds in regards to sexual activity and subsequent satisfaction, we originally hypothesized that romance (embodying one-on-one intimacy) might display a stronger association with cognitive status outcomes than sexual satisfaction. The data presented suggest otherwise, indicating that the act of romance must be sufficient in delivering sexual satisfaction if it is going to potentially contribute to the longitudinal maintenance of cognitive status in the aging population. Further work understanding the nuanced relationship between romance and sexual satisfaction in the aging population is clearly needed.

Previous research has linked a greater degree of social support to lower rates of cognitive decline in older adults (Barnes et al., 2004), however, social support was not predictive of cognitive decline in these data. Notably, social support was high for respondents who remained cognitively unimpaired as well as those who declined, indicating high levels of social support overall. The degree of social support reported may have been influenced by the fact that all respondents were married at baseline (Evans et al., 2019). These findings indicate the importance of the relationship between sexual satisfaction and cognitive impairment irrespective of social support.

Similarly, participants’ beliefs about sexuality lacked predictive value for longitudinal maintenance of cognitive status in the present study. The beliefs about sexuality subscale assessed for two distinct concepts: one’s beliefs or values about sexuality overall as well as one’s beliefs about their own and their spouse’s desire and ability to engage in a sexual relationship. The lack of statistical difference in scores may be a result of (1) a cohort effect lending to similar attitudes toward sex and sexuality, and (2) similar levels of objective and subjective health, indicating comparable physical ability to engage in sexual relationships.

In this sample, age and sex were unrelated to future cognitive decline. However, higher educational attainment was associated with a lower likelihood of cognitive decline, which is consistent with recent literature linking higher education to delayed onset of accelerated cognitive decline (Clouston et al., 2020). A higher percentage of respondents who lost a spouse went on to receive a diagnosis of MCI or dementia compared to those who did not experience a spousal death, however the difference was marginal, and lacked predictive ability in the model. As past research that suggests that experiencing a spousal death hastens cognitive decline in older adults (Shin, et al., 2018), future research in this area may improve evidence-based interventions for clinicians to support older adults after the death of a spouse.

Limitations

Information regarding respondents’ relationship status at the time of completing the intimacy survey was limited to married or not married. Due to the heterogeneity of intimate relationships of unmarried couples, we limited the sample to married older adults which greatly reduced the sample size. Furthermore, our sample was comprised of individuals who were able and willing to participate in longitudinal research and discuss sex and intimacy. It is likely that the sample was biased by those that feel most comfortable discussing issues of sexuality and intimacy, although we remain in equipoise as to whether this would potentially increase the likelihood of participation for those that wish to express their sexual dissatisfaction, or would enhance participation among those that want to share the success of this aspect of their lives. Future research should attempt to address such issues in larger and if possible, population-based samples.

Information regarding gender identity and sexual orientation was not available. Same-sex marriage was not legalized in Kentucky until 2015, and so we assume that all individuals included in this analysis were in opposite-sex marriages. Findings presented here offer groundwork for replication of this work in same-sex marriages, as well as diversity in sexual orientation. We fully recognize the importance of capturing additional variables in the future. Respondents were predominantly White and highly educated, thus limiting the generalizability of the findings. We suspect there may be many nuances in the areas of sexuality and intimacy that are highly dependent on sociocultural experiences and norms that the present study could not possibly hope to address. With this being said, the homogeneity of our sample provides a starting point for discussion and discovery that we hope will be extended to other populations with diverse socioeconomic, racial/ethnic, gender identity, and sexual preference backgrounds and orientations. Lastly, our response rate of 46.4% may introduce a self-section response bias in findings, although we note that age and gender were matched between respondent and non-respondent groups providing some degree of validation in this select cohort, whereas only educational attainment between respondents and non-respondents differed in our sample. Future studies exploring intimacy in relation to cognitive health that collect relationship status information, as well as sexual orientation and gender identity among diverse populations, are clearly needed.

Conclusion

This study is the first to demonstrate that intimacy is related to future cognitive decline over a 10-year span. Specifically, our findings support a potentilly predictive association of sexual satisfaction on longitudinal cognitive health status. Mental health and medical professionals may wish to consider the opportunity to improve care and longitudinal outcomes by addressing issues of sexual satisfaction in older adults without cognitive impairment.

Table 1.

Summary of items within subscales of the interpersonal relations, intimacy, and sexuality survey

Romance

I am able to share my thoughts with my spouse

I participate in social activities with my spouse outside of home

I give or receive a hug daily

My partner and I still hold hands

My partner and I still kiss

My partner and I kiss or hold hands in public

I am satisfied with the number of kisses I receive from my partner

Beliefs about Sexuality

People should not engage in public displays of affection

I am physically able to participate in a sexual relationship

I have a strong desire to be sexually active

My partner is still physically able to participate in a sexual relationship

My partner is still interested in sexual activity

Sexual activity is a critical part of a good relationship

Sex becomes less important to people as they age

Sexual Satisfaction

I would like to give more kisses to my partner each day

I am satisfied with the way my sexual needs are being met

I am worried about the sexual aspects of my life

I am disappointed in the quality of my sex life

My sexual relationship is good compared to most

I feel sad when I think about my sexual experience

Social Support

I have someone I can share my thoughts and feelings with

Others share their thoughts and feelings with me

I talk to my children(ren) or other family members at least once a week

I feel that my emotional needs are being met in these interactions

I participate in social activities with others outside my home

Clinical implications.

• Sexual satisfaction may represent a novel social determinant of cognitive health

• Enhanced surveillance of sexual satisfaction may help identify those at increased risk for future cognitive status changes including the development of MCI and or dementia

• Clinicians working with older adults should have a low threshold for referral to couples or sex therapists if signs and symptoms of sexual dissatisfaction are identified or suspected.