Assignment

The Impact of Aging on Human Sexual Activity and Sexual Desire

Osmo Kontula Population Research Institute, Helsinki, Finland

Elina Haavio-Mannila Department of Sociology, University of Helsinki

This article empirically studies how much aging modifies human sexual activity and sexual desire, and what the most important determinants in this change are. The analyses are based on 2 representative national sex surveys conducted in Finland in the 1990s. As a result of female widowhood, aging men had a higher incidence of sexual intercourse compared with aging women; and in relationships, women were more likely than men to report lack of sexual desire. In regression analysis, age was a predictor of sexual activity but not of sexual desire, when controlling for the impact of other factors. Relationship duration did not play an impor- tant role in sexual activity or sexual desire when controlling for a number of other variables. Sexual desire, valuing sexuality, and a healthy partner were important to female sexual activity; and high sexual self-esteem, good health, and active sexual history were important to male sexual activity. To keep up their sexual desire, both men and women needed good health, good sexual functioning, positive sexual self-esteem, and a sexually skilful partner.

The Impact of Aging on Sexual Activity

Sexuality and the desire for intimacy are essential and important human features from birth until death. Gerontologists and other medical experts generally agree that continued sexual interest and activity can be therapeutic for older men and women (Willert & Semans, 2000). Aging women who have a partner with whom they can enjoy intimacy are in better mental health compared with women without this kind of partner. For most elderly people, sexual activity is still an important means for expressing love and caring (Campbell & Huff, 1995).

Numerous studies have identified a decrease in sexual interest and activity in old age (Araujo, Mohr, & McKinlay, 2004; Beutel, Schumacher, Weidner, & Brähler, 2002; Blanker et al., 2001; Dennerstein & Lehert, 2004; Hayes & Dennerstein, 2005; Laumann et al., 2005; Lindau et al., 2007; Marsiglio & Donnelly, 1991; Nicolosi et al., 2006). This transition has been explained, for example, by biological and psychological factors, diseases, mental conditions, boredom with the relation- ship, and widowhood (DeLamater & Sill, 2005; Thienhaus, 1988). However, seldom have these factors

been comprehensively studied by representative national survey samples.

In analyzing the data, it is necessary to distinguish between the role and impact of aging as such and mem- bership in a specific generation. Aging plays an impor- tant role in sexuality, but the generations subsequent to the era of sexual revolution of the 1960s and 1970s have considerably modified their sexual values and behavior patterns. Carpenter, Nathanson, and Kim (2007) found that lower levels of satisfaction in older age stemmed from generational differences rather than from aging per se. This transition will apparently continue in subsequent generations.

Biological factors have an indisputable effect on sex- ual activity among the elderly, but they do not directly determine it. Numerous studies have shown that as men age, they undergo adaptations in physiology, hor- monal levels, and sensory capacities that reduce, on average, human sexual desire and activity (Bartlik & Goldstein, 2001; Bortz, Wallace, & Wiley, 1999). Chronic disorders such as cardiovascular disease, hyper- tension, diabetes, arthritis, and prostate disease may have a negative effect on sexual functioning and response (DeLamater & Sill, 2005). There are also psychological impacts of illnesses and surgical interven- tions on sexual functioning.

DeLamater and Sill (2005) believed that biological factors provide a necessary, but not sufficient, condition

Correspondence should be addressed to Osmo Kontula, Popu-

lation Research Institute, Family Federation of Finland, Kalevankatu

16 B1, First Floor, P.O. Box 849, Helsinki, FIN-00101, Finland.

E-mail: [email protected]

JOURNAL OF SEX RESEARCH, 46(1), 46–56, 2009

Copyright # The Society for the Scientific Study of Sexuality

ISSN: 0022-4499 print=1559-8519 online

DOI: 10.1080/00224490802624414

for sexual functioning. To come up with a comprehensive approach to aging sexuality, a bio-psychological perspec- tive is needed—one that combines biological, psychologi- cal, and socio-environmental factors and realms.

In the psychological realm, sexuality includes identity, body image, self-esteem, eroticism, emotions, and their expression and imagination (Badeau, 1995). Good quality sexual intimacy is only achievable by individuals who are mature, independent, have good self-esteem and trust, and respect their partners; in short, those who have the capacity for emotional intimacy (Kingsberg, 2002).

In the social realm, Trudel, Turgeon, and Piché (2000) listed a number of social determinants and fac- tors that influence sexual activity. These include social taboos; conjugal status; and knowledge about sexuality, self-esteem, and attitudes toward sexuality. In other words, the expressions of sexuality in a given society are governed also by common ideas, norms, and regula- tions. In the moral and religious realms, sexuality embodies learned individual values (Badeau, 1995).

In the social realm, the availability of a partner definitely influences sexual expression, as does the qual- ity of interaction and communication with that partner. Aging can be related to sexual desire and activity through the social influences that are a result of the longer duration of relationships. If a long-term relation- ship has been unsatisfactory, a new relationship may have a positive impact on sexual functioning.

Some people may take their emotional and sexual relationship for granted and become lazy and inattentive in their sexual activities. Lifelong psychological conflicts may come to the fore as couples age (Beutel et al., 2002). External factors, such as the needs of aging parents or grandchildren, may draw emotional energy away from a couple (Kingsberg, 2000). Many older, widowed women drift into a state of sexual disinterest. The death of a spouse often leads to a cessation of sexual activity. By turning off their interest in something they do not have and see little likelihood of getting, a widow may prevent herself from becoming frustrated or depressed (DeLamater & Sill, 2005).

The Impact of Aging on Sexual Desire

and Sexual Dysfunction

Kingsberg (2002) argued that sexual desire refers to people’s interest in being sexual and is determined by the interaction of three related, but separate, compo- nents: (a) drive, (b) beliefs and values, and (c) motivation. Hartmann, Philippsohn, Heiser, and Rüffer-Hesse (2004) added to this conceptual approach by offering that sexual desire is facilitated by engagement in sexual situations with physical and subjective feelings of arousal. Experi- encing pleasure triggers arousal, which subsequently triggers desire. Arousal and desire coexist and reinforce each other.

Sexual dysfunction is the inability to react emotion- ally or physically to sexual stimulation in a way expected of the average healthy person or according to people’s own standards. Sexual dysfunction may affect various stages in the sexual response cycle—desire, excitement, and orgasm (Kingsberg, 2002). In the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text revision; American Psychiatric Association, 2000), sex- ual problems are classified into four major areas: (a) sexual desire disorders, (b) sexual arousal disorders, (c) orgasmic disorders, and (d) sexual pain disorders (Malatesta, 2007).

There have been numerous attempts to measure the determinants of decreased sexual interest. Bartlik and Goldstein (2001) found that in retirement, diminished income, divorce, unresolved anger, separation from loved ones, medical illness, major depression, and the use of certain medications can lead to diminished sexual desire. Hartmann et al. (2004) reported that decreased sexual interest was related to self-reported negative emotional and psychological feelings (e.g., lower self- esteem, insecurity, and loss of femininity).

DeLamater and Sill (2005) found that having a partner was a significant predictor of desire among women but not among men. Their results suggested that a woman’s desire is attuned to her relationship context, but a man’s desire is not. Among men, sexual problems tend to be more associated with physical health and aging than among women (Laumann et al., 2005).

According to Laumann et al.’s (2005) global study, older age consistently increased the likelihood of most sexual problems among men but not among women. Only lubrication problems (problems with sexual desire were not measured) were positively associated with female older age.

This article has two principal aims:

1. To describe how relationships, sexual desire, fre- quency of sexual intercourse, and quality of sex evolve, on average, as men and women age, and as the duration of a relationship increases.

2. To analyze the determinants or predictors of the variation in frequency of sexual intercourse and sexual desire among aging men and women with a special focus on whether age as such has a sig- nificant effect on sexuality in the aging popu- lation, even after controlling for the effects of other relevant predictors.

Method

Participants

Two national sex surveys were conducted in Finland in 1992 and 1999. The data collection was financed by the Academy of Finland (1992) and the Finnish

THE IMPACT OF AGING ON SEXUAL ACTIVITY

47

Ministry of Social Affairs and Health (1999). No com- mercial or interest groups provided financial support for the studies. The samples were drawn at random from national population registers. They were representative of the total population within the age range (18–74 years) chosen for this study.

The sample in 1992 was 2,250. The 1992 survey was conducted by trained interviewers from Statistics Finland (the government’s central office of statistics). The researchers participated in the training of the inter- viewers. Most of the interviewers were middle-aged women. The respondents were sent a letter before an interviewer made personal contact with them.

To reduce the potential influence of taboos against open discussions of sexual matters, the titles of the stu- dies indicated that they also covered other aspects of life in addition to sexuality. The 1992 Finnish study was entitled, ‘‘A National Survey of Human Relations, Sex- ual Attitudes and Lifestyles.’’ In 1992, 2,250 Finns (aged 18–74 years) completed face-to-face interviews during which each interviewee also completed, in privacy, a self-administered questionnaire that covered the intimate questions. The response rate was 76%.

In 1999, a mail survey with mostly the same questions was conducted in Finland. The sample drawn from a national population register was 3,269; 1,496 responses were received from persons aged 18 to 81, corresponding to a response rate of only 46%. To compare the results with the first survey, people older than 74 years are excluded here. The data analyzed here is from 1,432 respondents in the 1999 survey. The lower number of respondents in 1999 compared to the survey in 1992 was due to the lower response rate in 1999.

In this study, the respondents of these two national sex surveys were combined in the age group of 45 to 74 years. There were altogether 1,549 respondents— 705 men and 844 women (Table 1). This combination of surveys was conducted to get more statistical power into the analyses.

By analyzing the distributions of several identical, retrospective questions measuring sexual issues in differ- ent generations, Kontula (2001) was able to show that the low response rate in 1999 did not have a major impact on the sexual history-derived results of people under the age of 55 and of women over the age of 55. In the 55- to 74-year-old age group, male respondents had their sexual initiation older, were more faithful, and had less sexual partners in their lifetime compared with the corresponding birth cohort interviewed in 1992. These differences were statistically significant. In sum, these men were more monogamous. Due to this selection bias, some results of the sexual activities among aging men over 55 years are low-bound estimates.

More detailed information on the sampling, interview- ing, and questionnaires is available in Haavio-Mannila and Kontula (2003); Haavio-Mannila, Kontula, and Kuusi (2001); and Kontula and Haavio-Mannila (1995).

The questionnaire closely followed that used in a survey by Sievers, Koskelainen, and Leppo (1974), but some new questions were added to the survey conducted in 1992.

The dependent variables are sexual intercourse and sexual desire. Sexual intercourse is measured on the basis of the following question: ‘‘How often have you had sexual intercourse during the last month (30 days)?’’ The eight possible responses range from 1 (not at all) to 8 (every day or more frequently). In the questionnaire, intercourse is defined as ‘‘sexual interaction in the form of vaginal intercourse, oral sex or stimulation by hand.’’ In this article, other forms of sexual activity (e.g., mas- turbation) are not discussed. Contrary to some popular assumptions, other types of sexual activities do not appreciably substitute sexual intercourse as people grow older (Haavio-Mannila & Kontula, 2003).

Sexual desire was measured by a sub-question to the following question: ‘‘In the last year, have you had any of the following problems in your sexual life—lack of own sexual desire?’’ The four possible responses range from 1 (not at all) to 4 (very frequently). The responses to the questions measuring the independent variables have been published in the books mentioned earlier.

The results are analyzed by cross-tabulating and using correlation and linear regression techniques. First, comparisons are made between men and women in the three age groups: 45 to 54, 55 to 64, and 65 to 74. In the text, the 10-year age ranges are often referred to as ‘‘50,’’ which means ages 45 to 54; ‘‘60,’’ meaning ages 55 to 64; and ‘‘70,’’ meaning ages 65 to 74. Second, people are divided according to the duration of their present couple relationship. The number of respondents in each category is presented in Tables 1–3.

In the cross-tables, differences between gender and age group, gender and duration of relationship groups, and the interaction between gender and the two other variables are tested by using the general linear model. The statistical significances are presented in the tables as p values.

The assumed predictors of sexual intercourse and desire refer to the following areas of social and sexual life: social background and lifestyle, type of present relation- ship, quality of relationship, sexual functioning, health, sexual self-esteem, sexual attitudes, and sexual satisfac- tion. Variables that have a statistically significant corre- lation with the two dependent variables among men, women, or both are included in a linear regression analy- sis explaining the frequency of intercourse and desire.

Altogether, 28 variables correlating with sexual inter- course or desire were included in the regression analysis. They are as follows:

Social background:

� Age � Years of education � Income

KONTULA AND HAAVIO-MANNILA

48

Relationship:

� Duration of current relationship � Type of relationship � Touching and physical closeness in the relationship

� Considers sex important for the happiness of the relationship

Sexual intercourse and satisfaction:

� Number of sexual partners in lifetime

Table 1. Social and Sexual Characteristics of Men and Women in Three Age Groups

Men, Age Group Women, Age Group Significance

Characteristics

45–54

(n¼330) 55–64

(n¼223) 65–74

(n¼152) All

(N¼705) 45–54

(n¼365) 55–64

(n¼268) 65–74

(n¼221) All

(N¼844) Age

Group Gender Interaction

Type of couple relationship

Marriage 69 70 73 70 70 63 46 62 .005 .000 .000

Cohabitation 8 3 2 5 7 4 5 5 .000 ns ns

Living apart together 8 9 8 8 10 7 3 7 ns ns ns

Single 15 18 17 17 13 26 46 26 .000 .000 .000

Total 100 100 100 100 100 100 100 100

Sexual partners

Number of sexual partners

in lifetime

Mean 17 18 9 16 5 3 3 4 .007 .000 .002

Median 7 5 4 5 3 2 2 2

At least two sexual partners

in last year

26 13 7 18 8 4 5 6 .000 .000 .002

Sexual intercourse

Sexual intercourse in a week 64 45 23 50 49 25 10 32 .000 .000 ns

No sexual intercourse

in last year

5 13 24 11 10 25 52 25 .000 .000 .000

Latest intercourse very or

quite pleasurable

90 90 81 88 82 61 53 68 .000 .000 .000

Orgasm in latest intercourse 94 92 89 92 66 53 56 60 .002 .00 ns

Sexual desire and functioning

Lack of own sexual desire caused

problems very or quite

often in last year

13 12 30 15 32 54 54 42 .000 .000 .000

Erection difficulties quite

often in last year

8 16 30 14 12 28 33 20 .000 .011 ns

Lubrication difficulties quite

often in last year

6 15 24 12 13 36 31 23 .000 .000 .015

Health

Own illness has caused problems in

sexual interaction very or

quite often in last year

5 14 32 12 4 13 19 9 .000 .016 ns

Good mental health: Few

stress symptomsa

(mean, range 10–27)

23.7 23.8 23.7 23.7 22.3 22.4 22.8 22.4 ns .000 ns

Self-rated sexual self-esteem

Considers himself or herself as sexually:

Skilful 41 37 36 39 36 26 28 31 .037 .006 ns

Active 62 58 52 59 41 29 33 36 .010 .000 ns

Attractive 26 27 26 27 32 31 30 31 ns ns ns

Sexual self-steem sum-scale meanb 9.9 9.6 9.1 9.7 9.2 8.1 8.3 8.7 .000 .000 ns

Sexual attitudes

Regards sex without love as right 53 42 28 44 35 24 17 27 .000 .000 ns

Regards pornography as arousing 64 57 55 60 36 24 20 28 .000 .000 ns

Sexual satisfaction

Very or quite satisfied with sexual

life as a whole

82 79 77 80 80 70 75 76 .015 ns ns

Note. Values are percentages unless otherwise noted. a

A sum-scale of nine stress symptoms: insomnia, headache, tension or anxiety, overstrain, general feeling of fatigue or weakness, dizziness,

nightmares, trembling hands, and burning cheeks or sweating without physical exertion. The values given to each symptom are 1 (frequently), 2

(sometimes), and 3 (hardly ever). The range of the sum-scale is 10 to 27, and the reliability (Cronbach’s alpha) is .74. b The range of the sum-scale composed of the three variables is 3 to 15, and its reliability (Cronbach’s alpha) is .82.

THE IMPACT OF AGING ON SEXUAL ACTIVITY

49

� Frequency of sexual intercourse in the last 12 months

� Latest intercourse was pleasurable � Satisfied with sex life as a whole

Sexual desire:

� Seldom experiences lack of sexual desire � Seldom experiences lack of partner’s sexual desire

Sexual functioning:

� Erection difficulties experienced rarely � Lubrication difficulties experienced rarely � Partner seldom comes too quickly � Partner seldom comes too slowly � Seldom comes too slowly � Seldom comes too quickly � Own illness has seldom caused sexual problems � Partner’s illness has seldom caused sexual problems

Sexual self-esteem and attitudes:

� Self-rated sexual self-esteem � Regards pornography as arousing � Regards sexual intercourse without love as wrong

Lifestyle and mental health:

� Frequency of exercise � Frequency of alcohol use � Body mass index � Stress symptoms

Results

Development of Sexual Living Conditions Through

the Aging Process

The most important determinant of frequency of sexual intercourse among the aging population is having

Table 2. Characteristics of the Couple Relationship of Men and Women in Three Age Groups

Men, Age Group Women, Age Group Significance

Characteristics

45–54

(n¼282) 55–64

(n¼183) 65–74

(n¼128) All

(N¼593) 45–54

(n¼311) 55–64

(n¼196) 65–74

(n¼122) All

(N¼629) Age

Group Gender Interaction

Type of couple relationship

Marriage 81 86 88 84 81 85 87 83 .042 ns ns

Cohabitation 10 3 2 6 8 7 5 7 .005 ns ns

Living apart together 9 11 10 10 11 8 8 10 ns ns ns

Total 100 100 100 100 100 100 100 100

Duration (means, years) and quality

of the relationship

22 30 39 28 23 32 40 29 .000 .039 ns

Couple relationship is very

or fairly happy

81 82 86 83 79 79 72 78 ns .028 ns

Discussing sexual matters with partner

is not at all or not very difficult

87 83 83 85 82 78 75 80 ns .016 ns

There is a suitable amount of touching

and physical closeness in the

couple relationship

77 79 80 78 75 77 77 76 ns ns ns

Considers sexual life very or fairly

important for the happiness of

the relationship

82 88 69 82 75 59 52 66 .000 .000 .000

Satisfaction with frequency

of intercourse

Satisfied with the frequency

of intercourse

51 60 61 56 67 69 61 67 ns .016 ns

Prefers more frequent sexual

intercourse in present relationship

48 38 35 43 23 14 21 20 .007 .000 ns

Partner’s sexual problems and health

Lack of partner’s desire has caused

problems very or quite often

in last year

18 23 40 23 12 15 19 14 .001 .000 ns

Partner comes too quickly very or

quite often in last year

4 4 14 5 27 38 34 31 .015 .000 ns

Partner comes too slowly very or

quite often in last year

26 21 28 25 10 14 18 12 ns .000 ns

Partner’s illness has caused problems

in sexual interaction very

or quite often in last year

3 10 22 8 7 20 36 15 .000 .000 ns

Note. Values are percentages unless otherwise noted.

KONTULA AND HAAVIO-MANNILA

50

Table 3. Characteristics of Sexual Life of Men and Women in Three Groups According to Duration of the Relationship

Men, Duration in Years Women, Duration in Years Significance

Characteristics

0–19

(n¼126) 20–39

(n¼346) 40þ

(n¼118) 0–19

(n¼132) 20–39

(n¼346) 40þ

(n¼146) Duration Gender Interaction

Type of couple relationship

Marriage 37 98 98 31 97 100 .000 ns ns

Cohabitation 25 1 1 29 2 — .000 ns ns

Living apart together 38 1 1 40 1 — .000 ns ns

Quality of relationship

Mutual love in

the relationship

72 75 58 70 81 68 .000 ns ns

Couple relationship is

very or fairly happy

79 83 87 77 80 74 ns .023 ns

Discussing sexual matters with

partner is not at all or

not very difficult

90 83 83 80 81 74 ns .008 ns

There is a convenient amount of

touching in the couple relationship

80 79 76 79 73 75 ns ns ns

Considers sexual life very or fairly

important for the happiness

of the relationship

83 84 77 68 70 51 .000 .000 ns

Sexual partners

Number of sexual partners in lifetime

Mean 31 12 8 8 3 2 .000 .000 .000

Median 15 5 3 5 1.5 1

At least two sexual partners in last year 24 17 7 12 3 1 .000 .000 ns

Sexual intercourse

Sexual intercourse in last week 67 58 34 51 45 17 .000 .000 ns

No sexual intercourse in last year 2 4 14 4 6 26 .000 .003 ns

Satisfied with the frequency of intercourse 52 54 67 66 70 59 ns .041 .011

Would like to have intercourse

more frequently

47 44 30 26 18 18 .015 .000 ns

Latest intercourse very or quite pleasurable 91 89 88 78 78 49 .000 .000 .000

Orgasm in latest intercourse 94 94 88 69 63 47 .000 .000 ns

Sexual desire and functioning

Lack of own sexual desire caused

problems very or quite often in last year

15 12 25 31 40 59 .000 .000 ns

Lack of partners sexual desire has

caused problems very or quite

often in last year

15 24 32 16 12 19 .038 .005 ns

Erection difficulties quite often in last year 9 14 23 11 16 39 .000 .012 ns

Lubrication difficulties quite often in last year 7 12 22 20 18 44 .000 .000 .038

Partner comes too quickly very or quite often 5 6 6 34 26 48 .003 .000 .004

Partner comes too slowly very or quite often 19 26 29 10 9 25 .011 .001 ns

Health

Own illness has caused problems in

sexual interaction very or quite

often in last year

15 12 25 31 40 59 .000 .000 ns

Partner’s illness has caused problems

in sexual interaction very or

quite often in last year

6 5 18 9 12 32 .000 .001 ns

Mental health: Few stress

symptoms (mean, range 10–27)

22.7 24.1 23.9 22.1 22.4 22.7 .000 .002 .039

Self-rated sexual self-esteem

(mean, range 3–15)

10.4 9.8 8.7 9.8 8.7 7.9 .000 .000 ns

Liberal sexual attitudes

Regards sex without love as right 54 46 28 48 26 17 .000 .000 ns

Regards pornography as arousing 67 61 55 39 34 20 .000 .001 ns

Sexual satisfaction

Very or quite satisfied with

sexual life as a whole

84 84 81 83 79 76 ns ns ns

Note. Values are percentages unless otherwise noted.

THE IMPACT OF AGING ON SEXUAL ACTIVITY

51

a regular partner, usually a husband or wife. This has a number of implications when comparing how, on average, the aging population lives in its couple relation- ships and engages in sexual intercourse. Table 1 gives an overview of the actual social and living conditions in the three age groups (45–54, 55–64, and 65–74) for all survey respondents, including those with no partner. Table 2 compares only those respondents who had a partner at the time of the survey.

The crucial gender difference among the aging popu- lation is that women over the age of 60 were less likely than men to have a permanent sexual partner. At the age of 70, less than one half of the women had a husband or other regular partner. Among men in the oldest age group, the proportion living in a couple relationship was more than four-fifths.

The proportion of people in a relationship of mutual love decreased from the youngest age group to the oldest. Not all people with a regular partner lived in a relationship of mutual love. At the age of 70, more than one half of the men (57%), but less than one half of the women (41%), were in a loving relationship.

Frequency of sexual intercourse declined considerably with age. Because men were more likely to have a regular partner, they had a much higher frequency of intercourse compared with women. At the age of 70, one-fourth of the men and one-tenth of the women had intercourse at least once a week. These figures are somewhat higher than in the United States (Lindau et al., 2007). One-fourth of the men and one half of the women had not engaged in any acts of intercourse over the last year.

When the aging women and men had no steady partner, they very seldom had sexual intercourse. The sex lives of the oldest study participants were quite monogamous; only 7% of men and 5% of women reported having multiple partners (i.e., more than 1 sex- ual partner in the past year). Of men at the age of 50, one-fourth had had at least two partners in the last year.

The quality of intercourse (i.e., finding intercourse pleasurable) was almost the same in all three age groups among men, but women in their 70s enjoyed it less than younger women. Only one half of the oldest women con- sidered their latest intercourse pleasurable, and 37% had an orgasm. When controlling for the importance of sex, the slight negative relation between orgasm and older age disappeared.

Already at the age of 60, one half of the women reported somewhat frequent lack of sexual desire. Among men, the proportion was much lower: 15%. Men seemed to preserve their sexual interest more even as they aged.

Sexual dysfunction or problems increased as people aged but perhaps less than one might expect. Two-thirds of men and women reported no frequent problems in erection or lubrication (female arousal) even at the age of 70. One-third of men and one-fifth of women reported that an illness had disturbed their sexual life. Health was a relevant determinant of sexual activity in the aging

population, but most people lived without health issues disturbing their sexual life.

Sexual self-esteem did not change very much as people aged. The proportion of study participants who considered themselves sexually skilful and active decreased significantly with increasing age, but there was no age difference with regard to self-rated sexual attractiveness. It is probable that people used others their own age, not younger people, as the reference group in these estimations.

Attitudes toward certain sexual issues, such as sex without love and pornography, were more liberal among men compared with women, and among the younger age groups compared with the older.

People who had a regular sexual partner often reported that they considered their relationship to be happy and that the relationship had a sufficient amount of touching and physical closeness (Table 2). Among the older respondents, the quality of the relationship did not suffer, although sexual intercourse was less frequent. This makes sense because the proportion of respondents that considered sex important for the happiness of the relationship was also lower among the oldest group. At the age of 70, two-thirds of the men and one half of the women considered sex important for the happi- ness of their relationship. When people were divided into those who considered sex important and those who did not, the quality of the relationship was not associated with the frequency of sexual intercourse. This was most apparent among older women.

Women reported a lack of sexual desire on the part of a partner almost as frequently as men in couple relation- ships (table not shown here) reported experiencing lack of sexual desire—at a rate of 14% versus 15%, respect- ively. However, only 23% of men reported lack of sexual desire in their partner, whereas 42% of women in couple relationships reported experiencing lack of desire.

Sex in Relationships of Longer Duration

In this study, the respondents who were married, cohabiting, or living apart together (LAT; couples who live separately) were divided into three groups (�19, 20–39, and 40+ years) according to the number of years they had lived together (Table 3). The aim was to study possible differences in relationships that have had a shorter or longer duration. It has often been argued that in longer relationships, partners get bored with each other and that their desire for sexual intercourse with the partner decreases.

When comparing those participants who have been together over 40 years with those together for a shorter duration, there may have been some degree of bias e.g., those who have divorced or separated, possibly due to sexual problems, may have been over-represented in the shorter duration groups.

KONTULA AND HAAVIO-MANNILA

52

A high number of years lived together had no impact on the quality assessments of the relationships. Regard- less of the duration of the relationship, respondents viewed their relationship as happy, and their assessment of touching and physical closeness was the same in all three duration of relationship groups. This held true, although the feeling of mutual love was somewhat less common in the relationships that had lasted at least 40 years. Men considered sex more important for the happiness of the relationship than women, and they still held this view in longer relationships.

There was a drop in the frequency of sexual inter- course in relationships that had lasted at least 40 years, in which only one-third of men and one-sixth of women had had intercourse at least once a week in the last month. In the last year, 14% of men and 26% of women in the longest relationships had had no sexual inter- course in the last year. The reasons for this abstention were rooted in lack of sexual desire, as well as illness, including erectile dysfunction, causing problems in sexual performance.

Of men who had suffered from lack of sexual desire very frequently or somewhat frequently over the last year, 22%, and 20% of similar women, had had no sex- ual intercourse. Men and women with a higher level of sexual desire, at 4% and 14%, respectively, had led their life during the last year without any sexual intercourse. Abstention among people suffering from illnesses that had disturbed their sexual activity very frequently or somewhat frequently was 33% among men and 23% among women, whereas it was only 2% and 13%, respectively, among healthier participants.

These results resemble the findings of a U.S. survey where the most common reason for sexual inactivity was the male partner’s physical health and where about one-fourth of sexually active older adults with a sexual problem reported avoiding sex as a consequence (Lindau et al., 2007).

Men were able to keep the quality of intercourse very good even in relationships of long duration. They assessed their latest intercourse as pleasurable and had orgasms as frequently as men in relationships of shorter duration. Women had not been able to keep up this high quality of intercourse in relationships lasting over 40 years. Less than one half of such women considered their latest intercourse pleasurable, and less than one half had an orgasm. Men had more difficulties getting and main- taining an erection as they aged. This may have had unwanted implications for female sexual satisfaction.

Thirty-five percent of men in the longest relationships and one-fifth of corresponding women would have pre- ferred more frequent intercourse in their relationship. One-fifth of women would have preferred less frequent intercourse. Also, 18% of women in relationships lasting over 40 years would have liked to have sex more often. Aging women were a quite heterogeneous group in terms of sexual interest. Previously, 5% to 15% of women have

been found to improve their sexual function and activity with increasing age (Hayes & Dennerstein, 2005).

The relationships of longer duration often faced part- ners’ decreasing sexual desire, increasing difficulties in sexual functioning, and increasing rates of illness that caused sexual problems in their interaction with their part- ner. In particular, lack of female sexual desire increased greatly with the duration of the relationship. More than one half (59%) of women, who had been in their current relationship for at least 40 years, reported lack of sexual desire very frequently or somewhat frequently.

There was some decrease in male desire with the growing number of years in a relationship, but only one-fourth of the men in longer relationships reported this problem. Although this gender discrepancy in desire surely caused continuous problems in communication between the partners, the assessment of the happiness of the relationship did not change. Among the aging population, happiness in the couple relationship was not as dependent on sexual interplay between the partners as it was among younger couples.

Predictors of Sexual Intercourse and Sexual Desire

The determinants of frequency of sexual intercourse and desire were studied through regression analysis with the aim of having the results controlled for the influence of the other variables included in the multivariate analy- sis. This analysis was carried out separately for men and women. It includes only people in couple relationships because many questions were addressed only to them (see Table 3).

The following nine variables did not explain frequ- ency of intercourse and desire in the regression analyses: education, income, frequency of exercise, body mass index, stress symptoms, duration of the present relation- ship, preferring more frequent intercourse (or satisfac- tion with the frequency of it), partner seldom coming too quickly, and regarding sex without love to be wrong. Their correlations with the dependent variables disap- peared when controlling for the impact of other social and sexual factors.

The most important result of the regression analysis was that the effect of age on frequency of sexual intercourse and sexual desire was significantly weaker when intervening factors (listed in the Methods section) were taken into con- sideration (Table 4). Nevertheless, age was still strongly related to the frequency of sexual intercourse even after controlling for other variables, but the influence of age on sexual desire disappeared completely after the controls.

Relationship duration did not play an important role in the frequency of sexual intercourse or sexual desire among either gender after controlling for the effects of other variables in the regression model. Also, relation- ship duration lost its predictive role in sexual intercourse and sexual desire after these controls. There are other

THE IMPACT OF AGING ON SEXUAL ACTIVITY

53

intervening factors that explain why sexual activity and desire decline in longer relationships.

In addition to age, important predictors of frequency of sexual intercourse among aging men were high self-rated sexual self-esteem, good health, active and pleasant sexual history (high number of sexual partners, satisfied with sex life as a whole, latest intercourse pleasurable), frequent alcohol use, and considering sex important in the couple relationship. Among women, the predictors of active sexual intercourse were a high value placed on sex in the couple relationship, satisfac- tion in their sex life as a whole, a healthy partner, and a higher level of own sexual desire.

An ‘‘ideal’’ aging couple was a healthy man with high sexual self-esteem and an active sexual history who met a woman who considered sexuality important for the happiness of the couple relationship and who had a fairly high level of sexual desire. Both the man and the woman would also have had a history of satisfying sex- ual experiences. The outcome would be a couple engag- ing in a high frequency of intercourse into their old age.

As mentioned earlier, sexual desire was not influenced by age when controlling for the other factors affecting it in the regression analysis. To express this result simply, the negative effect of biological aging on sexual desire

disappears when people are satisfied with their sexual experience, are sexually functioning, and healthy.

Positive predictors of high sexual desire among both men and women were good health (own illness has not caused problems in sex life), good sexual functioning (erection and lubrication), positive sexual self-esteem, and a sexually skilful partner (partner is not too fast or too slow in sexual intercourse).

For men, but not for women, considering sexual life important, being satisfied with it, and finding pornogra- phy arousing increased the likelihood of high sexual desire. For women only, sexual desire was associated with frequent intercourse, enjoying their most recent intercourse, and with a LAT relationship. When women did not see their partners every day (because of separate addresses), they were probably more able to dream of the next meeting, helping to keep their sexual desire high.

Conclusion

Knowledge about sexuality among the aging popu- lation is now more important than ever before. In Europe, the proportion of people over 60 is expected to grow by 50% over the next 30 years, and women live

Table 4. Predictors of Frequency of Sexual Intercourse and Sexual Desire

Frequency of Sexual Intercourse Sexual Desire

Predictor Men a

Women b

Men a

Women b

Age, unadjusted �.36��� �.36��� �.16��� �.17��� Age, adjusted �.22��� �.21��� �.02 .04 Relationship

Lives apart together ns ns ns .11���

Touching and physical closeness in the relationship ns .12� ns ns Considers sexual life as important for the happiness

of the relationship

.10�� .21��� .11�� ns

Sexual intercourse and satisfaction

Number of sexual partners in lifetime .16��� .07 ns ns Frequency of sexual intercourse in last month ns ns ns .11��

Latest intercourse pleasurable .10�� ns ns .15��

Satisfied with sexual life as a whole .16��� .19�� .10�� ns Sexual desire

Seldom lack of sexual desire .08� .10�� ns ns Sexual functioning

Erection difficulties rare .08� .08� .15��� ns Lubrication difficulties rare ns ns ns .20���

Partner seldom comes too quickly ns ns ns ns

Partner seldom comes too slowly ns ns .18��� .08��

Seldom comes too quickly ns ns .13��� .16���

Own illness has seldom caused sexual problems .10��� ns .19��� .14���

Partner’s illness has seldom caused sexual problems ns .12�� ns ns Sexual self-esteem and attitudes

High self-rated sexual self-esteem (sum) .13��� ns .14��� .20���

Regards pornography as arousing ns ns .10�� ns Lifestyle and mental health

Frequent alcohol use .09�� ns ns ns Adjusted R2 (proportion of variance explained) .360 .364 .265 .357

Note. Values are standardized regression coefficients in linear regression analysis. aN¼588. bN¼624, �p < 0.05, ��p < 0.01, ���p < 0.001.

KONTULA AND HAAVIO-MANNILA

54

up to one-third of their lives post-menopausally (Kontula & Miettinen, 2005). As these numbers grow, some studies report decreasing trends in sexual activity and sexual desire among the aging population that is get- ting older every day (see the introduction of this article). We can dispute whether these findings are due to aging as such or if they can be explained, for example, through changes over generations or changes in social and health status that occur among the aging population.

In this article, we have identified and analyzed some psychological and social factors that may explain a sig- nificant proportion of the variation in sexual activity and sexual desire among the elderly. According to our results, age was a significant predictor of sexual activity even after controlling for the influence of other studied variables in the regression model. This is probably due to human adaptations in biological and physiological factors along the aging process, which have outcomes also in sexual activity. It is interesting to note that age did not predict sexual desire when controlling for other factors included in the study. This indicates that the basis of sexual motivation is not associated with biologi- cal age or the aging process per se.

Contrary to expectations, the duration of the relationship did not play an important role in sexual activity and sexual desire after controlling for the other factors in the study. In addition to age, the most impor- tant predictors of sexual activity among aging men were good health, high sexual self-esteem, a history of multiple partners, pleasurable previous experiences, and frequent alcohol use. For women, they were sexual desire, a healthy and skilful partner, valuing sexuality in marriage, and pleasurable previous experiences. Men were better able than women to keep up their sexual intercourse activity thanks to the low rates of widowed men among the age group.

In maintaining their sexual desire, both men and women combined psychological, social, and biological factors. Common predictors included good health (no chronic disorders), good sexual functioning, positive sexual self-esteem, and a skilful partner. In addition, fre- quent intercourse, previous pleasurable experiences, and living in a LAT relationship kept female desire going. For men, also pornography and valuing sexuality pre- dicted higher levels of desire. These were positive resources or factors that helped men and women keep their sexual motivation high during the aging process.

One interesting finding was that the oldest men and women were as likely as younger people to view their cou- ple relationship as happy, and were also equally happy with the physical closeness and touching that they experi- enced with their partner. The quality of the relationship had not suffered, although it had lasted more than 40 years, and sexual activities were less frequent than before. The decrease in sexual activity did not make them unhappy because they did not consider sex very important in their relationship.

For a great part of the aging population, sexuality continues to be an important and pleasurable part of everyday life. Bortz et al. (1999) built and defined a so-called ‘‘exemplar category for aging men.’’ It included men who reported no medical conditions, lived with a sexual partner, and had a positive perception of her (part- ner). Participants in the exemplar category were sexually active and satisfied. The exemplar group of men mani- fested high sexual function scores across the age range. Many of the men in our study met the very same criteria.

In previous literature, biological factors have been considered important in the decline of sexual activity among the aging. Of these factors, we could analyze only chronic illness (own or partner’s). This was one limitation of the study. Another limitation was that we were not able to measure and study the hormonal levels of our respondents.

Nevertheless, Dennerstein, Lehert, and Burger (2005) found that prior functioning and relationship factors are more important than hormonal determinants in female sexual activity in midlife. In addition, Hartmann et al. (2004) found that hormonal changes were relatively weak in their direct effects on well-being and sexuality, whereas the major factors affecting perimenopausal women’s sexu- ality were feelings for the partner, partner’s problems, subjective well-being, and number and severity of meno- pausal symptoms. Altogether, sexual experience seems to change by age less in terms of physiology and more in terms of emotions. This is also in line with our findings.

In future studies it would be valuable to conduct qualitative research into the reasons, motivations, and problems that lay behind sexual inactivity among aging married couples. In addition to bored and loveless rela- tionships, there are probably also marriages that suffer from serious physical illness, disability, and mental pro- blems including things such as grudges or desires to seek revenge for something that a partner has done in the past (e.g., unfaithfulness).

These sexuality issues are increasingly important also in nursing institutions and nursing homes. Generally speaking, studies have shown that nursing staff and resi- dent physicians claim to have positive attitudes toward the sexual activities of elderly people. However, these favorable attitudes toward sexuality have not necessarily been accompanied by corresponding behaviors (Trudell et al., 2000). We agree with Willert and Semans’s (2000) conclusions that nursing professionals should not be embarrassed about a resident’s sexual concerns, should provide information about sexual issues if asked, and should reassure residents with health problems that sexual expression is still possible.

References

American Psychiatric Association. (2000). Diagnostic and statistical

manual of mental disorders (4th ed., Text rev.). Washington,

DC: Author.

THE IMPACT OF AGING ON SEXUAL ACTIVITY

55

Araujo, A. B., Mohr, B. A., & McKinlay, J. B. (2004). Changes in

sexual function in middle-aged and older men: Longitudinal data

from the Massachusetts Male Aging Study. Journal of the

American Geriatrics Society, 52, 1502–1509.

Badeau, D. (1995). Illness, disability and sex in aging. Sexuality and

Disability, 13, 219–237.

Bartlik, B. & Goldstein, M. Z. (2001). Men’s sexual health after

midlife. Psychiatric Services, 52, 291–293, 306.

Beutel, M. E., Schumacher, J., Weidner, W., & Brähler, E. (2002).

Sexual activity, sexual and partnership satisfaction in ageing

men—Results from a German representative community study.

Andrologia, 34, 22–28.

Blanker, M. H., Bosch, J. L. H. R., Groeneveld, F. P. M. J., Bohnen,

A. M., Prins, A., Thomas, S., et al. (2001). Erectile and ejaculat-

ory dysfunction in a community-based sample of men 50 to 78

years old: Prevalence, concerns, and relation to sexual activity.

Urology, 57, 763–768.

Bortz, W. M., Wallace, D. H., & Wiley, D. (1999). Sexual function in

1202 aging males: Differentiating aspects. Journal of Gerontology:

Medical Sciences, 54A, M237–M241.

Campbell, J. M. & Huff, M. S. (1995). Sexuality in the older woman.

Gerontology & Geriatrics Education, 16, 71–81.

Carpenter, L. M., Nathanson, C. A., & Kim, Y. J. (2007). Physical

women, emotional men: Gender and sexual satisfaction in midlife.

Archives of Sexual Behavior, Retrieved September 13, 2007, from

http://www.springer.com/public+health/journal/10508.

DeLamater, J. D. & Sill, M. (2005). Sexual desire in later life. Journal

of Sex Research, 42, 138–149.

Dennerstein, L. & Lehert, P. (2004). Women’s sexual functioning, life-

style, mid-age, and menopause in 12 European countries. Meno-

pause, 11, 778–785.

Dennerstein, L., Lehert, P., & Burger, H. (2005). The relative effects of

hormones and relationship factors on sexual function of women

through the natural menopausal transition. Fertility and Sterility,

84, 174–180.

Haavio-Mannila, E. & Kontula, O. (2003). Sexual trends in the Baltic

Sea area (Series D41). Helsinki: The Population Research Insti-

tute, Family Federation of Finland.

Haavio-Mannila, E., Kontula, O., & Kuusi, E. (2001). Trends in sexual

life: Measured by national sex surveys in Finland in 1971, 1992 and

1999 and a comparison to a sex survey in St. Petersburg in 1996

(Working Papers E 10=2001). Helsinki: The Population Research

Institute, Family Federation of Finland.

Hartmann, U., Philippsohn, S., Heiser, K., & Rüffer-Hesse, C. (2004).

Low sexual desire in midlife and older women: Personality fac-

tors, psychosocial development, present sexuality. Menopause,

11, 726–740.

Hayes, R. & Dennerstein, L. (2005). The impact of aging on sexual

function and sexual dysfunction on women. Journal of Sexual

Medicine, 2, 317–330.

Kingsberg, S. A. (2000). The psychological impact of aging on

sexuality and relationships. Journal of Women’s Health &

Gender-Based Medicine, 9 (Suppl. 1), S33–S38.

Kingsberg, S. A. (2002). The impact of aging on sexual function in

women and their partners. Archives of Sexual Behavior, 31, 431–437.

Kontula, O. (2001, August). Response rate and selection bias in a sex

survey: An empirical test. Paper presented at the International

Union for the Scientific Study of Population XXIV General

Population Conference, Salvador, Brazil.

Kontula, O. & Haavio-Mannila, E. (1995). Sexual pleasures: Enhancement

of sex life in Finland, 1971–1992. Aldershot, Hampshire: Dartmouth.

Kontula, O. & Miettinen, A. (2005). Synthesis report on demographic

behaviour, existing population related policies and expectations

men and women have concerning the state (DIALOG Work Pack-

age 4, Report D15, Working Paper E19=2005). Helsinki: The

Population Research Institute, Family Federation of Finland.

Laumann, E. O., Nicolosi, A., Glasser, D. B., Paik, A., Gingell, C.,

Moreira, E., et al. (2005). Sexual problems among women and

men aged 40–80 y: Prevalence and correlates identified in the

Global Study of Sexual Attitudes and Behaviors. International

Journal of Impotence Research, 17, 39–57.

Lindau, S. T., Schumm, L. P., Laumann, E. O., Levinson, W.,

O’Muircheartaigh, C. A., & Waite, L. J. (2007). A study of sexu-

ality and health among older adults in the United States. The New

England Journal of Medicine, 23, 762–774.

Malatesta, V. J. (2007). Sexual problems, women and aging: An

overview. Journal of Women and Aging, 19, 139–154.

Marsiglio, W. & Donnelly, D. (1991). Sexual relations in later life: A natio-

nal study of married persons. Journal of Gerontology, 46, 338–344.

Nicolosi, A., Buvat, J., Glasser, D. B., Hartmann, U., Laumann, E. O.,

Gingell, C., et al. (2006). Sexual behaviour, sexual dysfunctions

and related help seeking patterns in middle-aged and elderly

Europeans: The global study of sexual attitudes and behaviors.

World Journal of Urology, 24, 423–428.

Sievers, K., Koskelainen, O., & Leppo, K. (1974). Suomalaisten suku-

puolielämä [The sexual life of Finns]. Helsinki, Finland: Werner

Söderstrom OY.

Thienhaus, O. J. (1988). Practical overview of sexual function and

advancing age. Geriatrics, 43, 63–67.

Trudel, G., Turgeon, L., & Piché, L. (2000). Marital and sexual aspects

of old age. Sexual and Relationship Therapy, 15, 381–406.

Willert, A. & Semans, M. (2000). Knowledge and attitudes about later

life sexuality: What clinicians need to know about helping the

elderly. Contemporary Family Therapy, 22, 415–435.

KONTULA AND HAAVIO-MANNILA

56