Skip Navigation


IJE Advance Access originally published online on June 11, 2007
International Journal of Epidemiology 2007 36(5):1136-1142; doi:10.1093/ije/dym109
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
36/5/1136    most recent
dym109v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Chen, H.-K.
Right arrow Articles by Chen, T. H.-H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, H.-K.
Right arrow Articles by Chen, T. H.-H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Published by Oxford University Press on behalf of the International Epidemiological Association © The Author 2007; all rights reserved.

A prospective cohort study on the effect of sexual activity, libido and widowhood on mortality among the elderly people: 14-year follow-up of 2453 elderly Taiwanese

Huang-Kuang Chen1, Chuen-Den Tseng2, Shwu-Chong Wu3, Ti-Kai Lee4 and Tony Hsiu-Hsi Chen1,5,*

1 Institute of Preventive Medicine, College of Public Health, National Taiwan University, Taipei, Taiwan.
2 Division of Cardiology, Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan.
3 Graduate Institute of Health Policy and Management, College of Public Health, National Taiwan University, Taipei, Taiwan.
4 Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan.
5 Division of Biostatistics, Graduate Institute of Epidemiology, College of Public Health, National Taiwan University.

* Corresponding author: Tony Hsiu-Hsi Chen, Division of Biostatistics, Graduate Institute of Epidemiology/Institute of Preventive Medicine/Centre of Biostatistics Consultation, College of Public Health, National Taiwan University, Room 521, No. 17, Hsuchow Road, Taipei, 100, Taiwan. E-mail: chenlin{at}ntu.edu.tw


   Abstract
Top
 Abstract
Introduction
 Material and methods
 Results
 Discussion
 Acknowledgement
 References
 
Background Sexual activity in elderly people is a topic of growing interest but the relationships of sexual activity, libido and widowhood to mortality have been barely investigated.

Methods A total of 2453 subjects enrolled from a nationwide survey on health status of residents aged 65 years or older in Taiwan between 1989 and 1991 were followed up until 31 December 2003 for ascertaining cause of death. Information on the frequency of sexual activity, libido (sexual desire), widowhood, disease status and relevant risk factors for risk of death at baseline were collected.

Results After controlling for age and relevant confounding factors, sexual activity was found to be inversely related to mortality {adjusted hazard ratio (aHR) = 0.67 [95% confidence interval (CI):0.56–0.80] for males, aHR = 0.84 (95% CI:0.65–1.09) for females and aHR = 0.72 (95% CI: 0.62–0.84) for both sexes combined}. Men having libido had lower mortality [aHR = 0.81 (95% CI:0.68–0.97)]. Widowhood status was positively correlated with mortality [aHR = 1.66 (95% CI: 1.25–2.19) for males, aHR = 1.33 (95% CI: 1.09 to –1.62) for females and aHR = 1.43 (95% CI: 1.21–1.68) for both sexes combined]. Sexual activity was also inversely related to mortality from stroke [aHR = 0.64 (95% CI: 0.41–1.00)].

Conclusions Sexual activity was associated with all-cause and cause-specific mortality independently of other risk factors. This finding was consistent with the elevated risk of death associated with widowhood for both men and women, and by the decreased mortality risk in men having libido.

Keywords Elderly, mortality, sexual activity


   Introduction
Top
 Abstract
 Introduction
Material and methods
 Results
 Discussion
 Acknowledgement
 References
 
With the growing numbers of older people and their longer life expectancy, the effect of sexual activity on mortality is of considerable interest. However, the association between sexual activity and mortality has been barely addressed and only reported in sporadic studies. Sexual activity was related to mortality in elderly people in a previous short-term follow-up study.1 Another study found sexual frigidity and dissatisfaction were associated with acute myocardial infarction.2 However, the credibility of these studies is weakened by a lack of prospective cohort design. In a longitudinal cohort study, Davey Smith et al.3 reported a protective effect of sexual activity on all-cause mortality in male aged 45–59 years.3 Whether such a protective effect can also be seen among elderly people is worthy of being investigated.

As sexual activity is part of psychosocial function it is also very interesting to investigate whether two attributes related to sexual activity, libido and widowhood, are associated with mortality. More importantly, since sexual activity may also vary with different cultures, the effect of sexual activity on mortality may also vary with ethnic/racial group. The relationships of sexual activity, libido and widowhood to mortality among oriental people are poorly understood.

By using a longitudinal cohort study with 14-year follow-up of 2543 elderly Taiwanese, the present study aimed to assess the association between sexual activity, libido and widow status and all-cause or specific-cause mortality after controlling for other important risk factors.


   Material and methods
Top
 Abstract
 Introduction
 Material and methods
Results
 Discussion
 Acknowledgement
 References
 
Study samples
The target population was people aged 65 years and older, dwelling in Taiwan between July 1989 and June 1991. The study design on the sampling frame and the enrolment of study subjects have been described elsewhere.4,5 In brief, the study cohort was based on the enrolment of 2600 random samples from a nation-wide survey covering four counties, including Taipei (Northern Taiwan), Taichung (Central Taiwan), Kaoshiung (Southern Taiwan) and Hualien (Eastern Taiwan). The sampling unit was based on the small administrative unit, called ‘Li’ in Taiwan. For each ‘Li’, two residents were randomly selected, including one man and one woman, respectively, from population registry by gender.

Study design
To accomplish the objectives mentioned earlier, those who had sexual activity were treated as the exposed group. A total of 2453 elderly people with complete information on sexual activity formed a prospective cohort and were followed up until 31 December 2003. The average time of follow-up year was 11 years. The primary outcome was based on all-cause or specific-cause of death that was ascertained through the linkage of this cohort to the Taiwan mortality registry.

Data collection
The underlying basic information from our cohort was collected by a questionnaire, physical examination and laboratory test. Some selected variables of interest are defined as follows.

  1. Sexual activity: people still had sexual intercourse with their sexual partner when enrolled into our cohort.
  2. Frequencies of sexual activity: low frequency: none or <1 per month, medium frequency: 1 per month ~1 per week and high frequency: >1 per week
  3. Libido: people with sexual desire
  4. Widowhood status: peoples’ spouse had died before they were enrolled into this study.
  5. Pre-existing disease: a general physical examination was also undertaken by physicians, covering a wide range of principal organs including skin, eyes, ear, tongue and so on. Subjects with any suspected disease were referred to a medical centre to undergo confirmatory diagnosis. We classified these into seven major categories, including cardiovascular diseases, diabetes mellitus, cerebrovascular disease, respiratory disease, disease of digestive system, disease of the urinary system and others.
  6. Others baseline variables: These included anthropometric measures (body mass index and blood pressure), life-style factors (such as smoking and alcohol drinking) and biochemical markers [such as glucose, albumin, globulin, total cholesterol, high-density lipoprotein (HDL), low density lipoprotein (LDL), triglyceride (TG), urea nitrogen (UN), creatinine (CRE) and uric acid (UA)]. Disability status was measured by Barthel Index (BI),6,7 which are widely used to assess functional outcome in patients with ischaemic stroke.

Statistical method
Study subjects were classified into two groups by presence of sexual activity, libido and widowhood status, and into three groups by frequencies of sexual activity (low frequency: none or <1 per month, medium frequency: 1 per month ~1 per week and high frequency: >1 per week). We computed cumulative survival curve to show the association between sexual activity (a binary variable dichotomized by Yes/No) and all-cause or specific-cause death by using Kaplan–Meier method. Adjusted hazard ratios (aHRs) either by age or by all possible confounding factors were also calculated by using Cox proportional hazards regression model. All the analyses were also re-analysed by the classification of frequencies of sexual activity into three groups as mentioned above.


   Results
Top
 Abstract
 Introduction
 Material and methods
 Results
Discussion
 Acknowledgement
 References
 
Risk factors associated with sexual activity are compared and presented in Table 1 by the stratification of gender. Risk factors associated with sexual activity included age, body mass index, diastolic blood pressure, smoking and medical history of having diabetes mellitus and stroke, and disability.

Figure 1 shows 14-year cumulative survival by having sexual activity. Subjects with sexual activity had higher survival rate than those in the absence of sexual activity (Logrank Formula = 66.839, P < 0.001). The difference in 14-year survival did not vary with gender.


Figure 1
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 Cumulative survival curve by whether to have sexual activity MSA (+): Men with sexual activity; MSA (–): Men without sexual activity; FSA (+): Women with sexual activity; FSA (–): Women without sexual activity

 

View this table:
[in this window]
[in a new window]

  		Table 1 Frequency, mean value and comparison of baseline characteristics by gender and sexual activity status

 
Table 2 shows the frequencies of having sexual activity, libido and being widowed at entry to the study. Males had higher likelihood of retaining sexual activity and having libido but lower proportion of staying in widowhood than females.


View this table:
[in this window]
[in a new window]

  		Table 2 Relationships of all-cause death and sexual activity, libido and widowhood

 
Table 2 also shows age-adjusted and fully aHRs for the association between sexual activity and all-cause death. With adjustment for age only, sexual activity was inversely related to risk for death; aHR = 0.61 (95% CI:0.51–0.72) for males, aHR = 0.68 (95% CI:0.53–0.88) for females and aHR = 0.71 (95% CI:0.62–0.81) for both sex combined.

After controlling for established risk factors for death as mentioned previously, the association between sexual activity and mortality was attenuated but similar negative associations still remained; aHR = 0.67 (95% CI:0.56–0.80) for males, aHR = 0.84 (95% CI:0.65–1.09) for females and aHR = 0.72 (95% CI:0.62–0.84) for both sexes combined.

The inverse relationship between libido and risk for death [aHR = 0.81 (95% CI: 0.68–0.97)] was found in males but not in females (Table 2). Table 2 also shows the positive association between widowhood and mortality. Overall mortality was greater for widowed people than the couple by 43% [aHR = 1.43 (95% CI: 1.21–1.68)] after controlling for other possible confounding variables.

Table 3 lists fully aHRs for various underlying causes of death. We found sexual activity was inversely related to mortality from stroke [aHR = 0.58 (95% CI: 0.37–0.89)]. The negative association still remained even when we excluded subjects with the pre-existing stroke [aHR = 0.64 (95% CI: 0.41–1.00)]. Controlling for other factors attenuated the associations for death from diabetes mellitus and coronary heart disease.


View this table:
[in this window]
[in a new window]

  		Table 3 Relationships of specific cause of death and sexual activity with and without including of subjects with previous specific disease history

 
Figure 2 shows the cumulative survival curves by frequency of sexual activity. Those in the medium and high-frequency groups had higher survivals than those in the low-frequency group (long-rank test, {chi}2 = 37.422; P < 0.0001) However, no substantial difference was noted between the high and medium-frequency groups.


Figure 2
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 Survival curve between different sexual activity frequency groups High: high frequency: >1 per week; Medium: medium frequency: 1 per month ~1 per week; Low: low frequency: none or <1 per month

 
Table 4 shows subjects with high frequency and medium frequency of sexual activity had lower risk for death by 13% [aHR = 0.87 (95% CI: 0.72–1.04)] and 23% [aHR = 0.77 (95% CI: 0.66–0.91)], respectively compared with those with low frequency after adjustment for other variables as included in Table 3. The small effect found in the high-frequency sexual activity group may be due to sparse subjects and deaths in this category as shown in Table 4.


View this table:
[in this window]
[in a new window]

  		Table 4 Estimated results of the hazard ratio of death of all causes among subjects with different sexual activity frequency

 

   Discussion
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
Acknowledgement
 References
 
Biosocial plausibility
By assessing sexual activity, libido and widowhood, the present study demonstrated an inverse association between sexual activity and risk for death in elderly subjects using a longitudinal cohort with 14 years of follow-up. We also found sexual activity was inversely related to mortality from stroke. In addition, it is very interesting to note that the elderly having medium frequency survived longer compared with those having low frequency. The association between libido and mortality was only seen in men. Widowhood was strongly associated with risk for death.

Few studies addressed the association between sexual activity and mortality. By using a cohort of 918 men aged 45–59 with 10 years of follow-up, Davey Smith et al.3 reported an inverse relation between the frequency of orgasm and mortality. Adjusted odds ratios for the risk of death was 1.9 (1.0–3.4) for the group with low frequency of orgasm compared with high frequency with adjustment for age and risk factors. A dose-response relationship between frequency of orgasm and risk for death was also demonstrated. The inverse relationship between sexual activity and mortality in the elderly people was consistent with the findings of Davey Smith et al. study. Our results together with Davey Smith's findings, both of which have longer follow-up, suggest that sexual activity may have a protective effect not only in middle-aged men but also in the elderly people.

Regarding specific causes of death, Ebrahim et al.,8 using the same cohort, reported a positive association between the frequency of sexual intercourse and fatal heart disease after controlling for age and risk factors with the order of adjusted relative risk equal to 2.80 (1.13–6.96) at 10 years of follow-up, which was higher than the corresponding figure after 20 years of follow-up. However, there was absence of association between the frequency of sexual intercourse and stroke. In contrast, we found an inverse association between sexual activity and stroke but not in coronary artery disease. The discrepancy may be attributed to different age groups include in each study.

Alternative explanation
Despite the fact that an inverse association between sexual activity and mortality reduction was demonstrated in this study, it may be speculated over whether such an association is subject to biases and alternative explanations. Information on sexual activity was self-reported which may have resulted in misclassification bias. However, as primary outcome is based on mortality, differential misclassification (i.e. higher frequency of orgasm with higher mortality) is unlikely. Non-differential misclassification may be possible, but it often yields a conservative result. In addition, as significant findings were also noted for widowhood, which is very unlikely to have misclassification, it is unlikely that information bias is responsible for the relation between sexual activity and mortality.

Although confounding factors have been well controlled in the multivariate analysis we cannot rule out the possibility of other uncontrolled confounding factors. However, we believe the impact of other factors may be trivial because age, disease status and disability at baseline together with other possible biological factors related to mortality were taken into account. In addition to age, of other confounding factors, disease status at baseline and disability were considered the most important. Since individual with disease at baseline may be more likely to die within 5 years after baseline we examined the impact of sexual activity on mortality by excluding deaths within 5 years after baseline. The age- aHRs were 0.67 (0.55–0.81) for males, 0.69 (0.59–0.92) for females and 0.75 (0.64–0.88) for both sex combined, which are similar to those estimated in Table 1. This suggests that sexual activity may be associated with the risk for death even if the effect of disease at baseline has been removed.

Disability defined by Barthel ADL index score is also one of important factors affecting sexual activity and mortality. As it has also been controlled in multivariable Cox regression model, the association between sexual activity and risk for death is independent of disability. Regarding libido, since disability has been included as one of confounding factors and the protective effect after controlling for other confounding factors still remained in males, we believe the independent association between libido and mortality may be possible. However, as libido may be also affected by other psychological health indicators a proxy measure of self-reported psychological health cannot be completely ruled out.

The associations between martial status and mortality have been reported in two large follow-up studies. Based on a large cohort of middle-aged men with 18 years of follow-up, Ben-Shlomo et al.9 reported all unmarried men (widowed, single and divorced/separated men) had higher total mortality than married men. By further considering specific cause of death and subgroups of unmarried men, widowed men were at increased risk of dying from ischaemic heart disease. Cancer mortality was elevated among divorced men. By following up 7735 men aged 40–59 years over 11.5 years, Ebrahim et al.10 found single men had higher risk for cardiovascular disease mortality as well as non-cancer and non-cardiovascular mortality after adjustment for potentially confounding variables. The positive relationship between widowed status and other non-cardiovascular disease mortality was observed in men. The risk for death was not elevated among divorced/separated men. In our study, widowhood status was positively associated with mortality in men and women. However, we cannot investigate the relationship of martial status to mortality as seen in two large cohort studies because our study aim was focused on sexual activity and the detailed information on martial status except widowhood status was not obtained. Moreover, previous studies on the relation between widowhood and mortality also revealed two major factors that affects mortality in widowhood status, namely age of losing their spouse and durations of bereavement.11–13 However, these two factors cannot be investigated in our study because of a lack of such information.

In conclusion, findings from the present study strongly suggest sexual activity was associated with mortality, particularly related to stroke, among the elderly people. A similar effect is also found in widowhood status for both sex and males who were reported to have libido.


   Acknowledgement
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgement
References
 
We are indebted to Dr Jane Warwick for her technical English editing.


KEY MESSAGES

  • Sexual activity is an independent factor associated with mortality reduction in elderly people.
  • The relationship of sexual activity to specific-cause if death rate was particularly strong for the cause of death attributed to stroke in elderly people.
  • The protective effect of libido on mortality reduction has been found in elderly men.
  • Widowhood status increased mortality rate in elderly people.

 


   References
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgement
 References
 
1 Persson G. Five-year mortality in a 70-year-old urban population in relation to psychiatric diagnosis, personality, sexuality and early parental death. Acta Psychiat Scand (1981) 64:244–53.[Web of Science][Medline]

2 Abramov LA. Sexual Life and Sexual Frigidity Among Women Developing Acute Myocardial Infarction. Psychosom Med (1976) 38:418–25.[Abstract/Free Full Text]

3 Davey Smith G, Frankel S, Yarnell J. Sex and death: are they related? Findings from the Caerphilly Cohort Study. Br Med J (1997) 315:1641–44.[Abstract/Free Full Text]

4 Wu TH, Lee TK, Yen MF, Tung TH, Chen TH. Long-term mortality assessment using biological measures among elderly people. Ten-year follow-up of 597 healthy elderly subjects in Taiwan. Family Pract (2002) 19:272–77.[Abstract/Free Full Text]

5 Lee TK, Huang ZS, Ng SK, et al. Impact of alcohol consumption and cigarette smoking on stroke among the elderly in Taiwan. Stroke (1995) 26:790–94.[Abstract/Free Full Text]

6 Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. MA State Med J (1965) 14:61–65.

7 Granger CV, Dewis LS, Peters NC, Sherwood CC, Barrett JE. Stroke rehabilitation: analysis of repeated Barthel index measures. Arch Phys Med Rehab (1979) 60:14–17.[Web of Science][Medline]

8 Ebrahim S, May M, Ben Shlomo Y, McCarron P, Frankel S, Yarnell J, Davey Smith G. Sexual intercourse ans risk of ischaemic stroke and coronary heart disease: the Caerphilly study. J Epidemiol Community Health (2002) 56:99–102.[Abstract/Free Full Text]

9 Ebrahim S, Wannamethee G, McCallum A, Walker M, Shaper AG. Marital status, change in marital status, and mortality in middle-aged British men. Am J Epidemiol (1995) 142:834–42.[Abstract/Free Full Text]

10 Ben-Shlomo Y, Davey Smith G, Shipley M, et al. Magnitude and causes of mortality differences between married and unmarried men. J Epidemiol Community Health (1993) 47:200–5.[Abstract/Free Full Text]

11 Nagata C, Takatsuka N, Shimizu H. The impact of changes in marital status on the mortality of elderly Japanese. Ann Epidemiol (2003) 13:218–22.[CrossRef][Web of Science][Medline]

12 Martikainen P, Valkonen T. Mortality after death of spouse in relation to duration of bereavement in Finland. J Epidemiol Community Health (1996) 50:264–68.[Abstract/Free Full Text]

13 Martikainen P, Valkonen T. Mortality after the death of a spouse: rates and causes of death in a large Finnish cohort. Am J Public Health (1996) 86:1087–93.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
36/5/1136    most recent
dym109v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Chen, H.-K.
Right arrow Articles by Chen, T. H.-H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, H.-K.
Right arrow Articles by Chen, T. H.-H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?