Diabetes mellitus (DM) is a common chronic condition in the United States with an estimated lifetime risk of 32.8% in men and 38.5% in women.1 Among men, diabetes is a recognized risk factor for sexual dysfunction with prior research documenting an over threefold increased risk of erectile dysfunction in diabetic compared with nondiabetic men.2 Among women, the effect of diabetes on sexual function is poorly understood with very little research examining whether rates of sexual activity or sexual dysfunction differ in diabetic compared with nondiabetic women or identifying risk factors for sexual dysfunction in diabetic women.3
Diabetes has the potential to affect sexual function in women through a variety of mechanisms, including vascular changes in the urogenital tissues affecting genital lubrication and neuropathy-mediated alterations in genital arousal response. Women's interest in, satisfaction with, and ability to participate in sexual activity may be influenced globally by the effect of diabetes on their overall health, physical and mental functioning, and interpersonal relationships.4 Additionally, sexual function may be adversely affected by diabetes medications or other health interventions directed at monitoring or treating this chronic disease.5 , 6
To examine the relationship of diabetes to sexual function in middle-aged and older women, we evaluated sexual activity, desire, satisfaction, and problems in a racially and ethnically diverse, population-based cohort of 2,270 women aged 40–80 years, including 486 women with diabetes. Among diabetic women, we also examined associations between diabetic medication use, end-organ complications, and other markers of disease severity to sexual activity and function.
MATERIALS AND METHODS
We conducted an ancillary study to the Reproductive Risks of Incontinence Study at Kaiser 2, a cross-sectional cohort study of risk factors for urinary tract dysfunction in middle-aged and older women. Between January 2003 and January 2008, women were recruited from Kaiser Permanente Northern California, an integrated health care delivery system serving approximately 25–30% of the northern California population. To be eligible for this cohort, women had to be between the ages of 40 and 80 years, to have been enrolled in Kaiser since age 24 years, and to have had at least half their childbirths at a Kaiser facility but were not required to have any symptoms or a history of genitourinary dysfunction.7 Women of nonwhite race or ethnicity were recruited to achieve a target race and ethnicity composition of 20% African American, 20% Latina, 20% Asian, and 40% non-Latina white.
To achieve a diabetes prevalence of at least 20% in the overall cohort, women were also oversampled from the Northern California Kaiser Permanente Diabetes Registry, a database of Kaiser Permanente Northern California patients that is updated annually through active surveillance of pharmacy, laboratory, and medical records. Prior studies have indicated that the registry has a sensitivity of 96% and a false-positive rate of 2%.8 Women who self-reported as having diabetes but were not listed in the diabetes registry were still classified as being diabetic if they met the following criteria normally used for registry inclusion: 1) use of a diabetes glycemic control medication; or 2) fasting blood sugar of 126 mg/dL or greater in the Kaiser Permanente Northern California database. Although autoantibody data were not collected systematically to provide a definitive determination of type 1 compared with type 2 DM, only seven participants (fewer than 2% of all diabetic women) reported that they were diagnosed with diabetes before the age of 30 years and started on insulin at the time of diagnosis, suggesting that the vast majority of participants had type 2 DM. The final Reproductive Risks of Incontinence Study at Kaiser 2 cohort consisted of 2,270 women aged 40–80 years, including 486 with diabetes.
Demographic characteristics, medical and gynecologic history, medication use, and health-related habits were assessed in all participants using self-administered questionnaires as well as in-person interviews conducted either at a Kaiser Permanente Northern California clinic or in participants' homes. Height and weight were measured by trained personnel at study visits for calculation of body mass index (BMI, calculated as weight (kg)/[height (m)]2). Diabetic participants were also asked to contribute blood samples for measurement of serum creatinine and hemoglobin A1c (HbA1c).
Among participants with diabetes, specific diabetic end-organ complications were identified using questionnaire measures, physical examination, laboratory studies, or all of these. Specifically, heart disease and stroke were assessed by asking women, “Has a doctor, nurse, or other health care provider ever told you that you have any of the following conditions: 1) Heart attack (myocardial infarction), angina, or other heart disease? 2) Stroke?” Peripheral neuropathy was assessed using the validated Michigan Neuropathy Screening Instrument, which incorporates both self-reported symptom data and interviewer-administered physical examination of the lower extremities (foot inspection, vibration sensation, reflex testing, and monofilament testing); scores of 2 or greater are indicative of peripheral neuropathy.9 Renal function was examined by estimating glomerular filtration rate from serum creatinine levels and participant age and weight using the Cockcroft-Gault equation; participants with a glomerular filtration rate less than 90 were considered to have at least stage 1 renal dysfunction.
Sexual activity and function were assessed using structured-item measures (see the Appendix at http://links.lww.com/AOG/A307) derived from the validated Female Sexual Function Index10 and previously administered in other large women's health studies.11 – 13 To ensure confidentiality, participants completed questions in private and submitted them to study personnel in sealed envelopes at their study visit. For this study, sexual activity was defined inclusively as “any activity that is arousing to you, including masturbation.” Women were first asked to indicate whether they had any sexual activity in the past 3 months and, if so, the frequency of that activity.
Participants' sexual desire or interest, overall sexual satisfaction, and sexual problems were assessed through structured Female Sexual Function Index items that were adapted to assess sexual function in the 3 months before each visit. Women's level of sexual desire and overall sexual satisfaction were assessed in all participants regardless of sexual activity, whereas specific sexual problems (ie, low level of arousal, difficulty with lubrication, difficulty achieving orgasm, and pain or discomfort during vaginal intercourse) were assessed only among women who reported some sexual activity in the past 3 months. To assess women's perception of the effect of their physical health on sexual function, all participants were additionally asked “How much has your physical health limited your sexual activity?” with response options ranging from “not at all” to “extremely.”
For the purposes of analysis, participants were categorized into one of three diabetes status groups based on whether they had diabetes and, if so, whether they were using insulin: 1) insulin-treated diabetic women; 2) non–insulin-treated diabetic women; and 3) nondiabetic women. These categories were chosen a priori based on the recognition that insulin use is a widely recognized indicator of diabetes severity and also represents a higher level of disease management burden that can interfere with day-to-day functioning and quality of life. Differences in the demographic and clinical characteristics of participants in these three categories were examined using χ2 tests for categorical variables and analysis of covariance for continuous variables. Next, we described the distribution of less than monthly sexual activity, less than moderate sexual desire, and less than moderate sexual satisfaction among women in each diabetes status category. Among women reporting at least some sexual activity in the past 3 months, the prevalence of specific sexual problems such as low or very low arousal, at least moderate difficulty with lubrication, at least moderate difficulty with orgasm, or at least moderate pain with vaginal intercourse was also examined among women in each diabetes status category. Differences in the distribution of these sexual function outcomes among women in different diabetes status categories were examined using χ2 tests.
Our initial multivariable logistic regression models compared sexual function outcomes among: 1) insulin-treated diabetic compared with nondiabetic women; and 2) non–insulin-treated diabetic compared with nondiabetic women adjusting for a core set of other factors with potential to influence sexual function (ie, age, race and ethnicity, marital or relationship status, menopausal status, history of sex with men or women, BMI, hysterectomy and oophorectomy, selective serotonin reuptake inhibitor [SSRI] use, and estrogen use). Although models examining frequency of sexual activity, desire, and satisfaction included all women, models examining specific sexual problems were confined to sexually active women and additionally controlled for frequency of sexual activity.
Subsequent analyses used multivariable logistic regression to examine independent associations between diabetes-related end-organ complications and sexual function outcomes in all diabetic participants, again controlling for potential confounders. Sexual activity, desire, and satisfaction outcomes were examined in all diabetic participants, whereas specific problems with lubrication, arousal, orgasm, or pain were examined in sexually active diabetic women only. Finally, multivariable logistic regression models were developed to examine relationships between HbA1c and sexual function outcomes adjusting for potential confounders. In these analyses, women with HbA1c levels 6.0–6.9, 7–7.9, and 8.0 or greater were compared with women with levels less than 6.0 as the reference group. All analyses were performed using SAS 9.1 statistical software. All study procedures were approved the institutional review boards of both the University of California San Francisco and the Kaiser Foundation Research Institute.
Of the 2,270 participants, 139 (6.1%) were insulin-treated diabetic, 347 (15.3%) were non–insulin-treated diabetic, and 1,784 (78.6%) were nondiabetic women (Table 1). Mean (±standard deviation) age was 55 (±9.2) years, 1,006 (44.4%) were non-Latina white, 443 (19.5%) were African American, 401 (17.7%) were Latina, 401 (17.7%) were Asian, and 18 (0.8%) were Native American. Age, race or ethnicity, marital or relationship status, parity, oophorectomy history, oral glycemic medication use, SSRI and estrogen use, alcohol use, and BMI differed significantly by diabetes status (Table 1). Of the diabetic participants, insulin-treated women tended to have more heart disease and peripheral neuropathy as well as higher HbA1c levels compared with non–insulin-treated women.
Overall, 63.7% of participants reported some sexual activity in the past 3 months. Of the 807 women who reported no sexual activity in the past 3 months, 271 (33.6%) indicated that lack of a partner and 224 (27.7%) indicated that partner health problems contributed to their sexual inactivity. The proportion of sexually inactive women reporting partner-related issues did not differ by diabetes status (P for heterogeneity >.05).
Insulin-treated diabetic women were less likely to report at least monthly sexual activity compared with either non–insulin-treated diabetic women or nondiabetic women (Table 2). Insulin-treated diabetic women were also more likely to report low sexual desire and satisfaction compared with non–insulin-treated diabetic women or nondiabetic women. Among sexually active participants, problems with lubrication were also more common in insulin-treated diabetic women compared with nondiabetic women (Table 2).
In multivariable analysis adjusting for age, race, marital or relationship status, history of sex with men, women, or both, parity, menopausal status, BMI, hysterectomy, oophorectomy, SSRI use, and estrogen use, the odds of reporting low overall sexual satisfaction were over twofold higher in insulin-treated diabetic women and over 40% higher in non–insulin-treated diabetic women compared with nondiabetic women (Table 3). However, no significant differences in sexual desire or frequency of sexual activity by diabetes status were observed.
Among sexually active women, insulin-treated diabetic women were also more than twice as likely to report difficulty with lubrication and 80% more likely to report difficulty achieving orgasm compared with nondiabetic women after adjusting for the same demographic and clinical factors (Table 3). No significant associations between diabetes status and other types of sexual problems (difficulty with arousal or pain or discomfort during intercourse) were detected in multivariable analyses.
When asked if their physical health limited their sexual activity, insulin-treated diabetic women were more likely than nondiabetic women to report that their health limited their sexual activity “quite a bit” or “extremely” in multivariable analysis (odds ratio [OR] 2.29, 95% confidence interval [CI] 1.49–3.51). However, non–insulin-treated diabetic women were not substantially more likely than nondiabetic women to feel that their health limited their sexual activity (OR 1.29, 95% CI 0.92–1.78).
Among all diabetic women (n=486), the most common diabetic end-organ complication was peripheral neuropathy as measured by Michigan Neuropathy Screening Instrument score (60.9%) followed by renal dysfunction (39.5%), heart disease (13.4%), and stroke (6.4%). In multivariable analyses, diabetic women were more likely to report less than monthly sexual activity if they had heart disease, renal dysfunction, or peripheral neuropathy (Table 4). Diabetic women with a history of stroke were more likely to report low overall sexual satisfaction than those without a stroke history. Diabetic women with peripheral neuropathy were also more likely to report less than monthly sexual activity, lower sexual desire, and limitation of sexual activity by physical health compared with those without neuropathy. Among sexually active diabetic women, no significant associations between specific diabetic end-organ complications and sexual problems such as difficulty with arousal, lubrication, orgasm, or pain with intercourse were observed in adjusted models (Table 4). There were also no significant associations between number of years since diabetes diagnosis and sexual function after adjustment for end-organ complications (P>.05 for all).
Of the diabetic participants, 62 (13.0%) had HbA1c levels less than 6.0%, 159 (33.4%) had HbA1c levels of 6.0–6.9%, 135 (28.4%) had HbA1c levels of 7.0–7.9%, and 120 (25.2%) had HbA1c levels of 8.0% or higher. In multivariable analyses, diabetic women with HbA1c levels of 8.0% or higher were less likely to report low sexual satisfaction (OR 0.36, 95% CI 0.16–0.80) compared with diabetic women with HbA1c levels less than 6.0%. No other associations between HbA1c and sexual function were detected.
In this cohort of ethnically diverse middle-aged and older women, diabetic and nondiabetic women reported similar levels of sexual desire and frequency of sexual activity after adjustment for other demographic and clinical factors. However, both insulin-treated and non–insulin-treated diabetic women were significantly more likely to report low overall sexual satisfaction compared with nondiabetic women, and problems with lubrication and orgasm were more common among insulin-treated diabetic women compared with nondiabetic women. These findings suggest that although many diabetic women are interested and engaged in sexual activity, diabetes is associated with a markedly decreased sexual quality of life in women either through complications of the disease itself or through use of treatments.
Our study also found that diabetic women with end-organ complications such as peripheral neuropathy, renal dysfunction, stroke, and heart disease were more likely to report decreased sexual activity or lower sexual satisfaction than diabetic women without these complications. These findings suggest that diabetic end-organ complications may play an important role in decreasing women's sexual quality of life and that raise the possibility that prevention of diabetic complications may be helpful in preventing sexual dysfunction in women with diabetes.
To date, there has been very limited study of the effect of diabetes on female sexual function with prior research tending to rely on small numbers of participants,14 – 17 lack nondiabetic control participants,16 use unidimensional measures of sexual function,16 , 18 or focus on referral or other narrow populations.15 – 21 Although a few previous studies have pointed to worse overall sexual function among diabetic women,15 , 20 , 22 our study underlines the importance of distinguishing between different aspects of female sexual function when evaluating the burden of this disease. Based on this research, diabetes and its complications appear to have a much greater effect on sexual problems such as lubrication and orgasm as opposed to sexual desire or subjective arousal. Furthermore, our study indicates that the adverse effects of diabetes on sexual function may be concentrated in women taking insulin, an apparently high-risk group for developing sexual problems.
One unexpected finding was that diabetic women with HbA1c levels of 8.0% or greater in this cohort tended to report higher overall sexual satisfaction compared with those with lower HbA1c independent of clinical characteristics such as BMI, medication use, or duration of diabetes. One possible explanation for this finding is that it reflects unmeasured differences in attitudes, expectations, or approaches to sexual activity among diabetic women with better compared with worse glycemic control. Diabetic women who were less motivated or interested in checking and controlling their blood sugars may have placed more priority on sexual activity, function, or both in their daily lives, resulting in higher reports of sexual satisfaction. Alternatively, diabetic women with worse glycemic control may have had lower expectations about sexual activity in the setting of their poorly controlled disease with the paradoxical result that they retained a stronger subjective sense of sexual satisfaction despite experiencing the same sexual difficulties. Differences in impulse control and other unmeasured personality factors could also have influenced both glycemic control and sexual satisfaction in diabetic participants.
This study benefits from a large and ethnically diverse sample of women, multidimensional measures of sexual function and problems, and assessment of a large number of disease-specific factors with the potential to influence sexual function, including laboratory data on HbA1c level and kidney function. However, several limitations to this research should be recognized. First, this was a cross-sectional study, and we were unable to examine longitudinal change in sexual activity and function over time or provide definitive evidence of causal relationships. Future studies should address whether poor diabetes control over time or de novo incidence of diabetic complications is associated with progression of sexual dysfunction in diabetic women. Second, although our measures were derived from previously validated questionnaires and have been used successfully in other women's health studies, they were adapted for the purposes of this research without being resubjected to detailed psychometric testing. Additionally, the majority of diabetic participants were believed to have type 2 DM, which is consistent with national statistics showing that the vast majority of adult diabetic patients have type 2 DM, especially in older populations. This may limit the generalizability of our findings to women with insulin-dependent DM. Lastly, some diabetic complications (heart disease, stroke) were assessed exclusively by self-report, and validation through clinical evaluation may be helpful to confirm these findings. Additionally, our power to detect associations between some diabetic complications (eg, stroke) and sexual function outcomes was partly limited by the relatively small number of events.
Several previous studies have suggested that psychological factors such as depression play a role in sexual dysfunction among diabetic women.15 , 20 , 22 – 25 Because detailed depression measures were not administered in our cohort, our study did not assess depression as a mediator of the effect of diabetes on female sexual function. However, our multivariable models did adjust for SSRI medication use given their known propensity to worse sexual function and found that relationships between diabetes and sexual function were independent of SSRI therapy.
In summary, in this large cohort of ethnically diverse women, we found that diabetic women did not differ significantly from nondiabetic women with regard to interest and engagement in sexual activity but did report lower levels of sexual satisfaction and more problems with lubrication and orgasm during sex, particularly if they were taking insulin. End-organ complications such as heart disease, stroke, neuropathy, and renal dysfunction were associated with decreased sexual activity or decreased sexual satisfaction among diabetic women. Based on this research, clinicians may want to consider actively assessing for sexual problems in diabetic women, particularly those taking insulin, and counsel diabetic women that prevention of end-organ complications may be important in preserving their sexual function.
1. Narayan KM, Boyle JP, Thompson TJ, Sorensen SW, Williamson DF. Lifetime risk for diabetes mellitus in the United States. JAMA 2003;290:1884–90.
2. Feldman HA, Goldstein I, Hatzichristou DG, Krane RJ, McKinlay JB. Impotence and its medical and psychosocial correlates: results of the Massachusetts Male Ageing Study. J Urol 1994;151:54–61.
3. Rutherford D, Collier A. Sexual dysfunction in women with diabetes mellitus. Gynecol Endocrinol 2005;21:189–92.
4. Lindau ST, Tang H, Gomero A, Vable A, Huang ES, Drum ML, et al.. Sexuality among middle-aged and older adults with diagnose and undiagnosed diabetes: a national, population-based study. Diabetes Care 2010;33:2202–10.
5. Enzlin P, Mathieu C, Vanderschueren D, Demyttenaere K. Diabetes mellitus and female sexuality: a review of 25 years' research. Diabet Med 1998;15:809–15.
6. Erol B, Tefekli A, Sanli O, Ziylan O, Armagan A, Kendirci M, et al.. Does sexual dysfunction correlate with deterioration of somatic sensory system in diabetic women? Int J Impot Res 2003;15:198–202.
7. Thom D, Van Den Eeden S, Ragins A, Wassel-Fyr C, Vittinghof E, Subak LL, et al.. Difference in prevalence of urinary incontinence by race/ethnicity. J Urol 2006;175:259–64.
8. Karter AJ, Ackerson LM, Darbinian JA, D'Agostino RB Jr, Ferrara A, Liu J, et al.. Self-monitoring of blood glucose levels and glycemic control: the Northern California Kaiser Permamente Diabetes registry. Am J Med 2001;111:1–9.
9. Moghtaderi A, Bakhshipour A, Rashidi H. Validation of Michigan neuropathy screening instrument for diabetic peripheral neuropathy. Clin Neurol Neurosurg 2006;108:477–81.
10. Rosen R, Brown C, Heiman J, Leiblum S, Meston C, Shabsigh R, et al.. The Female Sexual Function Index (FSFI): a Multidimensional Self-Report Instrument for the Assessment of Female Sexual Function. J Sex Marital Ther 2000;26:191–208.
11. Huang AJ, Stewart AL, Hernandez AL, Shen H, Subak LL; Program to Reduce Incontinence by Diet and Exercise. Sexual function among overweight and obese women with urinary incontinence in a randomized controlled trial of an intensive behavioral weight loss intervention. J Urol 2009;181:2235–42.
12. Subak LL, Wing R, West DS, Franklin F, Vittinghoff E, Creasman JM, et al.. Weight loss to treat urinary incontinence in overweight and obese women. N Engl J Med 2009;360:481–90.
13. Huang AJ, Subak LL, Thom DH, Van Den Eeden SK, Ragins AI, Kuppermann M, et al.. Sexual function and aging in racially and ethnically diverse women. J Am Geriatr Soc 2009;57:1362–8.
14. Erol B, Tefekli A, Ozbey I, Salman F, Dincag N, Kadioglu A, et al.. Sexual dysfunction in type II diabetic females: a comparative study. J Sex Marital Ther 2002;28(supp1 1):55–62.
15. Ogbera A, Chinenye S, Akinlade A, Eregie A, Awobusuyi J. Frequency and correlates of sexual dysfunction in women with diabetes mellitus. J Sex Med 2009;6:3401–6.
16. Ellenberg M. Sex and the female diabetic. Medical Aspects of Human Sexuality 1977;11:30–8.
17. Fatemi S, Taghavi S. Evaluation of sexual function in women with type II diabetes mellitus. Diab Vasc Dis Res 2009;6:38–9.
18. Kolodny RC. Sexual dysfunction in diabetic females. Diabetes 1971;20:557–9.
19. Abu A, Al Hajeri R, Khader Y, Shegem N, Ajlouni K. Sexual dysfunction in Jordanian diabetic women. Diabetes Care 2008;31:1580–1.
20. Nowosielski K, Drosdzol A, Sipinski A, Kowalczyk R, Skrzypulec V. Diabetes mellitus and sexuality: does it really matter? J Sex Med 2010;7:723–35.
21. Veronelli A, Mauri C, Zecchini B, Peca MG, Turri O, Valitutti MT, et al.. Sexual dysfunction is frequent in premenopausal women with diabetes, obesity and hypothyroid, and correlates with markers of increased cardiovascular risk. A preliminary report. J Sex Med 2009;6:1561–8.
22. Enzlin P, Mathieu C, Van den Bruel A, Bosteels J, Vanderschueren D, Demyttenaere K. Sexual dysfunction in women with type 1 diabetes: a controlled study. Diabetes Care 2002;25:672–7.
23. Wallner LP, Sarma AV, Kim C. Sexual functioning among women with and without diabetes in the Boston Area Community Health Study. J Sex Med 2010;7:881–7.
24. Enzlin P, Rosen R, Wiegel M, Brown J, Wessells H, Gatcomb P, et al.; DCCT/EDIC Research Group. Sexual dysfunction in women with type 1 diabetes: long-term findings from the DCCT/EDIC study cohort. Diabetes Care 2009;32:780–5.
25. Nowosielski K, Skrzypulec-Plinta V. Mediators of sexual functions in women with diabetes. J Sex Med 2011;8: 2532–45.