Abstract

Considerable research has been conducted into the effects of antihypertensive drugs on male sexual functioning. This remains underexplored in women, even though almost half of treated hypertensives are women. An ambulatory medical record-based, case-control study was designed to study sexual function in treated and untreated hypertensive women and healthy controls.

We conducted this study at a teaching hospital with satellite clinics in upstate New York. Of 3312 medical records reviewed, 640 premenopausal white women with or without mild hypertension (defined as blood pressure [BP] 140/90 and < 160/110 mmHg), in heterosexual relationships, with no other significant medical history, were eligible. Of these, 241 women agreed to participate, and 224 (35%) completed both a self-administered questionnaire and a telephone interview. Analysis was conducted on 211 women (107 healthy controls, and 104 mild hypertensives, of whom 37 were unmedicated and 67 medicated). Questions on sexuality were classified into seven composite variables and later further divided.

There were no demographic differences between participants and nonparticipants. Cases and controls differed only by age (P .01); therefore, subsequent analysis was age-adjusted. Current smokers reported a significantly lower mean score for orgasm than did nonsmokers (P= .04). Women with unmedicated and medicated hypertension did not differ significantly on sexuality scores and were subsequently combined. Using age-adjusted ANOVA, women with hypertension reported significantly decreased lubrication and orgasm and increased pain compared to nonhypertensive women. There were no significant differences by ANOVA in the quality of sexual functioning between six treatment groups.

In conclusion, the quality of female sexual functioning was quantified in an ambulatory outpatient setting. Hypertensive women, regardless of type of treatment, reported age-adjusted decrease in vaginal lubrication, less frequent orgasm, and more frequent pain when compared to nonhypertensive women. Emotional aspects of sexual functioning in hypertensive women do not appear to be impaired. These areas require further investigation. An incidental finding indicated diminished orgasm reported in current smokers, compared to nonsmokers, which was not associated with age or hypertension.

Considerable research has been conducted into the effects of antihypertensive drugs on male sexual functioning. This topic remains underexplored in women,1 even though almost half of treated hypertensives are women. Anecdotal reports about effects of older, noncardioselective drugs, however, indicate that women may be affected at rates similar to those found in men. Sexual side effects of medications and questions on sexuality are at times incorrectly extrapolated from men to women.2 Problems in women might be shown to exist if appropriate questions were asked.3

The paucity of research in women may in part be due to the difficulty in defining and evaluating objective parameters of their sexual functioning. Little is known about the frequency of diminished sexual quality among women in the general population;4–6 much less is known about diminished quality in women with chronic disease with and without the effects of pharmacotherapy.7–9 It is generally accepted, however, that the prevalence of sexual dysfunction in an ill population is probably higher than the estimated prevalence within a healthy population.10,11

With the exception of a small pharmacotherapeutic cross-over study12 and a large double-blind randomized controlled trial of older men and postmenopausal women,13 carefully controlled population-based studies on women that examine the association between hypertension, antihypertensive therapy, and quality of sexual functioning have not been done. Studies of this nature are the only valid means to measure incidence and prevalence of sexual dysfunction.10 We report the results of a case-control survey of premenopausal women participants in an upstate New York health care system that examined quality of sexual functioning and the effects of six hypertension treatment groups, including four major drug classes, combination therapy, and unmedicated hypertension.

Materials and methods

A review of 3312 medical records of women between 21 and 49 years of age resulted in 640 (19%) presumed premenopausal women who met the study criteria (in heterosexual relationships, with or without mild hypertension) and who were invited to participate in a cross-sectional survey evaluating sexuality and hypertension. The study took place between 1994 and 1996 in rural upstate New York. Of those invited, 241 (38%) agreed to participate and 224 (35%) completed both a self-administered questionnaire (SAQ) on sexuality (questionnaire available from authors) and a telephone interview on history of hypertension, general health, and medication usage, including over-the-counter preparations. These mailings occurred at three times in the year (spring, summer, fall), each time producing enough charts having been reviewed to justify a research assistant's time. The refusal rate may be attributed to the sensitivity of the subject, including questions dealing with the most intimate details of female sexuality.

Heterosexuality was initially defined by statements of marital status listed in the medical database at the beginning of the medical report. The medical record was reviewed for any notation indicating change in marital status. Participants were asked in both SAQ and interviewer questionnaire about their current marital/partner status. All women indicating involvement in a heterosexual relationship by questionnaire were invited to participate. Women in marriages were not analyzed separately from those not married but in active partnerships because of the focus of the study and the small numbers. Women in homosexual relationships were not selected because of the difficulty in obtaining an adequate sample size. Women who indicated lack of sexual activity were not included in the analysis. There is no information available on women not sexually active at time of interview, as they were excluded from analysis.

Participants were premenopausal, and were either in good health or had mild hypertension (blood pressure ≥ 140/90 mm Hg and < 160/110 mm Hg). Because of the homogeneity of the geographic area, only white women were studied. Only women with a diagnosis of essential, uncomplicated hypertension were included; women with secondary hypertension and hypertension diagnosed during pregnancy were excluded. Thirteen women discovered to be postmenopausal on interview were also excluded from analysis, leaving a study group of N = 211. A more detailed methodology is described elsewhere.14

The SAQ was designed based on questions identified in the literature,4,6,15–18 and included questions on socioeconomic status and social habits, quality of relationship, history of menstruation, pregnancy and menopause, and aspects of sexuality. A subsequent telephone interview elicited a more detailed medical history, including height, weight, and current medications. A carefully designed, well-controlled call-back methodology ensured a maximal completion rate.

Participants were stratified into one of three study groups: healthy controls, and medicated and unmedicated hypertensive cases. Healthy controls were selected by diagnostic related groups (DRG) from patients seen for routine obstetrical care, and hypertensive cases were selected if they had an International Classification of Diseases 9 (ICD 9) code for hypertension for any visit between January, 1991 and June, 1995.

Determination of hypertensive status was done in the following manner. Initial determination of hypertension, case versus healthy control, was made from information abstracted from the medical record, either a recorded diagnosis of hypertension, recorded treatment with standard antihypertensive therapy, or three recorded elevated blood pressures from consecutive visits. The telephone interviewer later asked specific questions about a diagnosis of hypertension, whether the hypertension was controlled with diet and exercise and/or medication, and listed all medications currently being taken. Subjects were classified according to the most current information on hypertensive status; hypertensive subjects were divided into either medicated or unmedicated hypertension (controlled by diet and exercise alone) on the basis of the telephone interviews.

Antihypertensive therapy was classified into one of the four drug classes commonly used as first-line monotherapy for hypertension: angiotensin converting enzyme (ACE) inhibitors, β-blockers, calcium channel blockers, and diuretics (thiazide and loop). Two additional classes were evaluated: women taking combination drug therapy, and those unmedicated.

χ2 analyses were used to compare cases and controls on sociodemographic and health variables. Each of the 47 questions on the emotional and physiologic aspects of sexuality was initially analyzed individually. These questions were then classified into seven composite variables8,19–21 to reduce the number of outcome variables (and therefore limit the number of statistical tests performed) and to simplify analysis. Four physiologic composite variables were desire/aversion, arousal/lubrication, attainment of orgasm, and physical pain (dyspareunia only was studied), and were derived from existing clinical definitions of female sexual dysfunction.19,22,23 The three remaining composite variables were based on emotional/relationship issues identified by prior research,8,19–21 and included quality of and satisfaction with the relationship, and frequency of sexual activity.

To account for unanswered questions, an individual's score for a specific composite variable was given as a mean response for the number of questions answered, rather than as the sum of the responses of the questions answered. Stem-and-leaf plots showed that, with the exception of pain (which showed some right skew), composites were grossly normally distributed. It should be noted that the nonparametric analysis did not rely on the normality of these distributions.

Results

Sociodemographic analysis

There were no significant demographic differences between 224 women who elected to participate in the study and 416 women who did not. Age and average blood pressure did not differ significantly between groups by t test (P = .21 for age, P = .20 for systolic blood pressure and .23 for diastolic blood pressure).14

Table 1 compares cases and controls on sociodemographic and health status variables. There were significant differences in age between the three groups: 34.8 years for 107 healthy controls, 37.9 years for 37 women with unmedicated hypertension, and 40.6 years for 67 women with medicated hypertension (P = 0). Because of the potentially confounding effects of menopause on sexuality, 13 women who had undergone natural and surgical menopause were deleted from the analysis. Healthy controls had a higher educational level (P < .01) and more children living at home (P < .01) than did either the unmedicated or medicated hypertensives. There was no significant difference between the three groups by current smoking status (P = .38), or by partner education level (P = .08, χ2 = 8.31) or cohabitation status (P = .07, χ2 = 8.72). Hypertensive women had a higher average body mass index (BMI) and average blood pressure (recorded from the medical record) (P < .01) than did healthy controls. There were significant differences between both average recorded blood pressures and self-reported blood pressures, with the latter being consistently lower (P < .01). This may reflect errors in self-reporting as well as a temporal lag between medical record abstracting and telephone interviews, during which time physiologic status may have changed.

Table 1.

Sociodemographic and health status by cases and controls (number/percent)

Sociodemographics Healthy Controls Unmedicated Hypertension Medicated Hypertension Statistics P Value 
Age, years (N = 107) 34.8 (N = 37) 37.9 (N = 67) 40.6 F = 38.30 .00* 
Education level      
Some high school 0/0.0% 1/2.7% 5/7.6% χ2 = 14.66 df 4 .00* 
High school/technical school graduate 21/19.6% 11/29.7% 23/34.3% 
College/postgraduate (13+) 86/80.4% 25/67.6% 39/58.2% 
Children at home      
Yes 102/95.3% 28/75.7% 51/78.5% χ2 = 14.54 df 2 .00* 
No 6/4.7% 9/24.3% 14/21.5% 
Health Status Healthy Controls Unmedicated Hypertension Medicated Hypertension Statistics P Value 
Current smoker      
Yes 27/25.2% 6/16.2% 19/28.4% χ2 = 1.93 df2 .38 
Body mass index (average) 25.35 30.57 32.73 F = 22.65 .00* 
Recorded × systolic BP 119.3 132.6 139.1 F = 69.50 .00* 
Recorded × systolic BP 73.8 87.6 88.7 F = 88.30 .00* 
Self-reported systolic BP 114.9 130.6 132.7 F = 25.75 .00* 
Self-reported diastolic BP 71.7 81.4 83.8 F = 28.50 .00* 
Sociodemographics Healthy Controls Unmedicated Hypertension Medicated Hypertension Statistics P Value 
Age, years (N = 107) 34.8 (N = 37) 37.9 (N = 67) 40.6 F = 38.30 .00* 
Education level      
Some high school 0/0.0% 1/2.7% 5/7.6% χ2 = 14.66 df 4 .00* 
High school/technical school graduate 21/19.6% 11/29.7% 23/34.3% 
College/postgraduate (13+) 86/80.4% 25/67.6% 39/58.2% 
Children at home      
Yes 102/95.3% 28/75.7% 51/78.5% χ2 = 14.54 df 2 .00* 
No 6/4.7% 9/24.3% 14/21.5% 
Health Status Healthy Controls Unmedicated Hypertension Medicated Hypertension Statistics P Value 
Current smoker      
Yes 27/25.2% 6/16.2% 19/28.4% χ2 = 1.93 df2 .38 
Body mass index (average) 25.35 30.57 32.73 F = 22.65 .00* 
Recorded × systolic BP 119.3 132.6 139.1 F = 69.50 .00* 
Recorded × systolic BP 73.8 87.6 88.7 F = 88.30 .00* 
Self-reported systolic BP 114.9 130.6 132.7 F = 25.75 .00* 
Self-reported diastolic BP 71.7 81.4 83.8 F = 28.50 .00* 
*

Denotes significance, P < .01.

BP = blood pressure.

Analysis of sociodemographic variables by sexuality variables

Sociodemographic variables were analyzed by the seven sexuality variables (Table 2) using nonparametric analysis of variance. College-educated women had a significantly increased mean score for frequency of intercourse (P = .02) when compared with women with a high school education, and women whose partners were college educated had significantly higher mean scores for desire, orgasm, and frequency (P ≤ .01). There were no significant differences in mean scores for sexual functioning for the presence of children at home or breast-feeding. Women who lived with a partner had a significantly higher mean score for quality of relationship than women who lived alone but who had a partner (P = .02); however, the latter group had significantly higher mean scores for desire and arousal than did the former (P = .02). Differences in sexual functioning by religion were not significant.

Table 2.

Mean scores for sociodemographic variables and smoking by seven composite variables on quality of sexual functioning

Sociodemographics Quality Desire Arousal Orgasm Pain Frequency Satisfaction 
Education level*        
 Some high school 3.96 3.64 2.91 3.56 4.69 2.63 3.4 
 High school/technical school graduate 4.12 3.55 2.64 3.83 4.45 2.85 3.64 
 College/postgraduate (13+) 4.16 3.68 2.91 4.00 4.61 3.02 3.71 
 P = .59 P = .72 P = .06 P = .10 P = .68 P = .02 P = .34 
Partner education level*        
 Some high school 3.86 3.31 2.6 3.81 4.4 2.65 3.5 
 High school/technical school graduate 4.12 3.61 2.86 3.8 4.5 2.88 3.67 
 College/postgraduate (13+) 4.22 3.78 2.89 4.13 4.65 3.08 3.78 
 P = .15 P = .01 P = .16 P = .00 P = .06 P = .00 P = .22 
Children at home        
 Yes 4.12 3.62 2.82 3.95 4.6 2.93 3.68 
 No 4.29 3.77 2.96 3.85 4.4 3.16 3.69 
 P = .30 P = .31 P = .30 P = .46 P = .12 P = .30 P = .84 
Breastfeeding        
 Yes 4.19 3.79 3.03 4.31 4.67 3.24 3.75 
 No 4.12 3.59 3.05 3.97 4.62 2.98 3.71 
 P = .62 P = .55 P = .51 P = .07 P = .96 P = .27 P = .79 
Cohabitation status*        
 Live with partner 4.16 3.65 2.83 3.95 4.57 2.95 3.7 
 Live alone, have partner 3.67 4.21 3.4 4.4 4.57 3.2 3.85 
 Live alone, no partner  3.3 2.7 3.62 4.64 3.06 3.26 
 P = .02 P = .02 P = .02 P = .19 P = .90 P = .45 P = .11 
Body mass index*        
 ≤ 27 91.29 107.34 106.59 109.86 102.24 108.33 108.47 
 > 27 105.46 101.82 105.45 99.54 104.66 102.88 101.75 
 P = .08 P = .51 P = .89 P = .22 P = .75 P = .51 P = .42 
Currently pregnant*        
 Yes 94.50 104.88 125.94 107.88 71.88 82.25 97.19 
 No 99.71 105.00 105.74 104.89 105.29 106.94 105.83 
 P = .80 P = 1.0 P = .36 P = .89 P = .09 P = .26 P = .70 
Type of birth control        
 Tubal ligation 94.44 100.14 81.98 94.53 92.75 83.05 93.10 
 Vasectomy 99.79 103.98 103.23 92.64 95.50 109.47 96.34 
 Natural 92.09 98.52 104.79 101.63 102.00 109.47 112.48 
 Hormonal 103.45 102.32 110.06 105.21 105.92 96.11 89.95 
 Barrier 81.75 91.21 102.42 103.88 97.98 111.69 99.37 
 IUD 107.31 110.31 150.25 129.06 145.50 116.56 141.88 
 P = .69 P = .95 P = .02 P = .62 P = .16 P = .10 P = .15 
Menopause*        
 Premenopause 95.62 101.85 105.30 101.21 100.20 103.78 100.99 
 Perimenopause 91.73 96.15 86.85 99.82 98.83 92.13 104.40 
 P = .72 P = .62 P = .11 P = .90 P = .90 P = .31 P = .77 
Breast tenderness        
 No 95.70 100.91 92.59 99.77 102.40 99.17 99.24 
 Before periods 96.22 110.75 120.83 109.97 95.37 107.30 113.01 
 All the time 77.94 71.60 92.55 56.55 109.90 103.80 97.85 
 Sometimes 110.15 95.13 127.60 117.30 107.23 115.33 88.53 
 P = .60 P = .23 P = .01 P = .04 P = .74 P = .69 P = .36 
Current smoker*        
 Yes 4.08 3.76 3.05 3.81 4.62 2.99 3.67 
 No 4.06 3.61 2.89 4.04 4.55 2.96 3.63 
 P = .79 P = .28 P = .23 P = .04 P = .64 P = .74 P = .56 
Sociodemographics Quality Desire Arousal Orgasm Pain Frequency Satisfaction 
Education level*        
 Some high school 3.96 3.64 2.91 3.56 4.69 2.63 3.4 
 High school/technical school graduate 4.12 3.55 2.64 3.83 4.45 2.85 3.64 
 College/postgraduate (13+) 4.16 3.68 2.91 4.00 4.61 3.02 3.71 
 P = .59 P = .72 P = .06 P = .10 P = .68 P = .02 P = .34 
Partner education level*        
 Some high school 3.86 3.31 2.6 3.81 4.4 2.65 3.5 
 High school/technical school graduate 4.12 3.61 2.86 3.8 4.5 2.88 3.67 
 College/postgraduate (13+) 4.22 3.78 2.89 4.13 4.65 3.08 3.78 
 P = .15 P = .01 P = .16 P = .00 P = .06 P = .00 P = .22 
Children at home        
 Yes 4.12 3.62 2.82 3.95 4.6 2.93 3.68 
 No 4.29 3.77 2.96 3.85 4.4 3.16 3.69 
 P = .30 P = .31 P = .30 P = .46 P = .12 P = .30 P = .84 
Breastfeeding        
 Yes 4.19 3.79 3.03 4.31 4.67 3.24 3.75 
 No 4.12 3.59 3.05 3.97 4.62 2.98 3.71 
 P = .62 P = .55 P = .51 P = .07 P = .96 P = .27 P = .79 
Cohabitation status*        
 Live with partner 4.16 3.65 2.83 3.95 4.57 2.95 3.7 
 Live alone, have partner 3.67 4.21 3.4 4.4 4.57 3.2 3.85 
 Live alone, no partner  3.3 2.7 3.62 4.64 3.06 3.26 
 P = .02 P = .02 P = .02 P = .19 P = .90 P = .45 P = .11 
Body mass index*        
 ≤ 27 91.29 107.34 106.59 109.86 102.24 108.33 108.47 
 > 27 105.46 101.82 105.45 99.54 104.66 102.88 101.75 
 P = .08 P = .51 P = .89 P = .22 P = .75 P = .51 P = .42 
Currently pregnant*        
 Yes 94.50 104.88 125.94 107.88 71.88 82.25 97.19 
 No 99.71 105.00 105.74 104.89 105.29 106.94 105.83 
 P = .80 P = 1.0 P = .36 P = .89 P = .09 P = .26 P = .70 
Type of birth control        
 Tubal ligation 94.44 100.14 81.98 94.53 92.75 83.05 93.10 
 Vasectomy 99.79 103.98 103.23 92.64 95.50 109.47 96.34 
 Natural 92.09 98.52 104.79 101.63 102.00 109.47 112.48 
 Hormonal 103.45 102.32 110.06 105.21 105.92 96.11 89.95 
 Barrier 81.75 91.21 102.42 103.88 97.98 111.69 99.37 
 IUD 107.31 110.31 150.25 129.06 145.50 116.56 141.88 
 P = .69 P = .95 P = .02 P = .62 P = .16 P = .10 P = .15 
Menopause*        
 Premenopause 95.62 101.85 105.30 101.21 100.20 103.78 100.99 
 Perimenopause 91.73 96.15 86.85 99.82 98.83 92.13 104.40 
 P = .72 P = .62 P = .11 P = .90 P = .90 P = .31 P = .77 
Breast tenderness        
 No 95.70 100.91 92.59 99.77 102.40 99.17 99.24 
 Before periods 96.22 110.75 120.83 109.97 95.37 107.30 113.01 
 All the time 77.94 71.60 92.55 56.55 109.90 103.80 97.85 
 Sometimes 110.15 95.13 127.60 117.30 107.23 115.33 88.53 
 P = .60 P = .23 P = .01 P = .04 P = .74 P = .69 P = .36 
Current smoker*        
 Yes 4.08 3.76 3.05 3.81 4.62 2.99 3.67 
 No 4.06 3.61 2.89 4.04 4.55 2.96 3.63 
 P = .79 P = .28 P = .23 P = .04 P = .64 P = .74 P = .56 
*

Kruskal-Wallis ANOVA.

Denotes significance.

Wilcoxon two-sample test.

IUD = intrauterine device.

Current smokers reported a significantly lower mean score for orgasm when compared to nonsmokers (P = .04). There was no significant difference by χ2 analysis in the distribution of smokers across the three study groups (χ2 = 1.93, P = .38), nor was there a significant difference by ANOVA in age between smokers (36.2 years) and nonsmokers (37.5 years), (P = .66).

Analysis of sexuality composite variables by cases/controls and antihypertensive therapy

Responses to individual questions comprising five of the composite variables (quality of relationship, desire, physical pain, frequency of intercourse, and sexual satisfaction) indicated no significant differences between cases and treated and untreated controls when initial χ2 analysis was performed. A sixth composite, orgasm, had one question (“In the last year, have you found you cannot reach orgasm/climax/come as easily as before?”) where healthy controls had a significantly better response than did medicated and unmedicated hypertensive women (χ2 = 11.14, P = .04). (Three questions that previously showed significant differences became nonsignificant when the 13 menopausal women were deleted from analysis.)

Table 3 shows mean composite variable scores and results for Kruskal-Wallis ANOVA for healthy controls and unmedicated and medicated hypertensive cases for each of the seven sexuality composite variables. Only the arousal composite variable had significantly lower mean scores for both medicated and unmedicated hypertensive women than for healthy controls (P = .05). The arousal composite variable contained three questions on lubrication, a physiologic response. Responses to two of the three questions were significantly worse for women with hypertension, whether or not they were taking medication: “Do you have vaginal dryness?” (P = .04); “Within the last year have you noticed increased vaginal dryness?” (P < .01). Only one of six questions on the emotional aspects of arousal showed a significantly worse response for hypertensive women: “In the last year, have you had more difficulty staying excited?” (P = .02). Arousal was subsequently divided into two separate composites, lubrication (physiologic) and arousal (psychologic) (Table 4).

Table 3.

Mean scores of composite variables and kruskal-wallis anova by seven variables (arousal and lubrication combined) for medicated and unmedicated hypertensive cases and healthy controls

 Quality Desire Arousal Orgasm Pain Frequency Satisfaction 
Healthy control 4.10 3.62 2.93 4.02 4.63 3.00 3.68 
Unmedicated hypertension 4.14 3.65 2.75 3.80 4.50 2.81 3.61 
Medicated hypertension 4.19 3.67 2.73 3.87 4.54 2.97 3.72 
 P = .99 P = .99 P = .05* P = .13 P = .27 P = .08 P = .40 
 Quality Desire Arousal Orgasm Pain Frequency Satisfaction 
Healthy control 4.10 3.62 2.93 4.02 4.63 3.00 3.68 
Unmedicated hypertension 4.14 3.65 2.75 3.80 4.50 2.81 3.61 
Medicated hypertension 4.19 3.67 2.73 3.87 4.54 2.97 3.72 
 P = .99 P = .99 P = .05* P = .13 P = .27 P = .08 P = .40 
*

denotes significance.

Table 4.

Continuous age-adjusted anova by eight variables (original arousal divided into lubrication and arousal)

 Quality Desire Lubrication Arousal Orgasm Pain Frequency Satisfaction 
F value 0.14 0.26 3.18 0.40 2.19 2.13 1.6 0.47 
Pr > F 0.87 0.77 0.04* 0.67 0.11 0.12 0.20 0.62 
 Quality Desire Lubrication Arousal Orgasm Pain Frequency Satisfaction 
F value 0.14 0.26 3.18 0.40 2.19 2.13 1.6 0.47 
Pr > F 0.87 0.77 0.04* 0.67 0.11 0.12 0.20 0.62 
*

Significant decrease in women with hypertension compared to healthy controls.

Additionally, as there were no significant differences between mean scores for women with unmedicated hypertension and women taking medication for hypertension (Table 3) (with the exception of frequency, which became a minor variable), these two groups were combined for further analysis. Because of concern that age could be a confounding variable, subsequent analysis was age-adjusted using parametric ANOVA (stem-and-leaf plots showed that each composite variable approximated a normal distribution, with the exception of pain). The lubrication composite indicated a significant decrease for women with hypertension compared to healthy controls (P = .04), whereas the difference in arousal was not significant (.67).

Table 5 shows univariate means and ANOVA results unadjusted and adjusted for age. The lubrication variable was significant both before (P < .01) and after (P = .04) age adjustment. Both orgasm (P = .11 before age adjustment) and pain (P = .17 before age adjustment) became significant after adjusting for age (P = .04). The pain variable should be interpreted cautiously, as there was right skew in the distribution of responses by stem-and-leaf plots and numbers were small.

Table 5.

Univariate means and anova (unadjusted and adjusted for age) for healthy and hypertensive women by eight sexuality variables

Univariate Means Quality Desire Lubrication Arousal Orgasm Pain Frequency Satisfaction 
Healthy (N = 107) 4.10 3.62 2.96 2.96 4.02 4.63 3.00 3.68 
Hypertension (N = 104) 4.18 3.67 2.66 2.81 3.86 4.53 2.92 3.69 
Non-age adjusted F value 0.66 0.29 7.01 2.41 2.59 1.86 1.14 0.00 
Pr > F 0.42 0.59 < 0.01* 0.12 0.11 0.17 0.29 0.95 
Continous age-adjusted F value 0.23 0.48 4.22 0.63 4.30 4.28 1.20 0.00 
Pr > F 0.63 0.49 0.04* 0.43 0.04* 0.04* 0.28 0.98 
Univariate Means Quality Desire Lubrication Arousal Orgasm Pain Frequency Satisfaction 
Healthy (N = 107) 4.10 3.62 2.96 2.96 4.02 4.63 3.00 3.68 
Hypertension (N = 104) 4.18 3.67 2.66 2.81 3.86 4.53 2.92 3.69 
Non-age adjusted F value 0.66 0.29 7.01 2.41 2.59 1.86 1.14 0.00 
Pr > F 0.42 0.59 < 0.01* 0.12 0.11 0.17 0.29 0.95 
Continous age-adjusted F value 0.23 0.48 4.22 0.63 4.30 4.28 1.20 0.00 
Pr > F 0.63 0.49 0.04* 0.43 0.04* 0.04* 0.28 0.98 
*

denotes significance.

Six hypertension treatment groups, including four monotherapy (ACE inhibitors, β-blockers, calcium channel blockers, and diuretics), combination therapy, and no drug therapy, were analyzed by the eight sexuality composite variables. There were no significant differences by ANOVA in the quality of sexual functioning between treatment groups both before and after continuous age adjustment. However, desire and frequency approached significance (P = .066 and P = .087, respectively).

Discussion

In a review of the literature in 1991, the Women's (caucus) Working Group on Women's Health of the Society of General Internal Medicine asked whether the paucity of data indicates no diminished quality of sexual function in women with hypertension, reflects inadequate methods of assessment, or indicates a lack of research on this topic.3 They concluded that when data on sexual function were available, either the questions were framed to refer only to male sexual functioning, or the investigators relied on the patients volunteering complaints, thus potentially underestimating the prevalence of dysfunction. The Working Group concluded that information does not exist because appropriate questions are rarely asked of women involved in clinical trials, which have historically excluded women, especially premenopausal women.24 The lack of reported data on diminished quality of sexual functioning appears to be due more to methodologic deficiencies and nonspecific questions in clinical trials than to a lack of cause and effect of hypertension and medication. A recent review of the literature since 1970 concluded that clinical pharmacotherapeutic trials involving hypertension and other chronic disease often neglect to take a detailed sexual history in female participants, infrequently study women only, and involve relatively few female investigators.25

Quality of sexual functioning is influenced by sex hormones, the emotional and physical health of the individual, and external factors such as availability and attractiveness of the sexual partner.7 Individuals may have diminished sexual quality that precedes medication use, or psychological problems, or other difficulties secondary to chronic disease that may be exacerbated by certain antihypertensive agents.7,9,10 In men, hypertension itself is associated with impaired sexual function, as are increasing age, smoking, and alcohol use.10,11 Given that certain areas of sexual physiology and reproductive embryology are similar in men and women, it follows logically that the sexual response in women may be impaired at a rate similar to that in men. Diminished quality of sexual functioning in women may manifest as orgasmic impairment, failure of vaginal lubrication, vaginismus, loss of libido, or infertility.1

A strength of this study lies with the fact that it surveyed healthy (except for cases with mild hypertension), sexually active, premenopausal women in an ambulatory outpatient setting, rather than in a clinic, and provided a continuum of sexual normalcy rather than a definition of dysfunction. Due to the cross-sectional design of the survey, and the questionnaire and interviewing process taking place in three separate contacts over the course of a year, seasonal variability in sexual activity and cyclical menstrual variability were assumed to be randomly distributed.12 The study also demonstrates that symptoms of sexual dysfunction can be elicited using a questionnaire in a community-based setting. Demographically, the selected sample did not differ substantially from the sampling frame and therefore is assumed to be representative of those eligible to participate in the study.

There were, however, limitations to the methodology. Nonresponse bias and self-selection are potential dangers in any study involving behavior, especially one with the emotional weight of sexuality. Nonparticipants required more reinforcement (follow-up) to elicit replies to the letter of invitation than did those who chose to participate, which raises the issue of potential reply bias among participants. Participants were not followed over time and were not asked to fill out daily diaries, a method that has previously been described as an effective means to evaluate sexuality,12 but one that may in itself lead to bias due to self-conscious observation.

There were few major sociodemographic differences between healthy and hypertensive women except for a significant age difference that was to be expected, given that prevalence of hypertension increases with age. After adjustment, however, age did not affect the study outcome. Hypertensive women taking medication had lower levels of education and higher BMI than did unmedicated hypertensives and healthy controls. However, there was no association between BMI and diminished sexual function, and only higher education was associated with increased frequency of intercourse.

A potentially important incidental finding was that mean score for orgasm in current smokers was significantly lower when compared to that in nonsmokers. This did not appear to be associated with age or hypertension. Cigarette smoking has been shown to cause diminished sexual functioning in men;10,11 this finding may also implicate smoking in diminishing ability to achieve orgasm.

Hypertensive women, regardless of type of treatment, reported an age-adjusted decrease in vaginal lubrication, less frequent orgasm, and more frequent pain when compared to women without hypertension. It is possible that increased pain and decreased orgasm are a function of diminished lubrication. Hypertensive women should be asked about vaginal dryness, as this problem may be easily remedied by use of commercially available lubricating agents before intercourse, or other means.

We conclude that premenopausal white women with hypertension may be at risk for diminished physiologic response that may lead to sexual dysfunction. Because erection and early lubrication occur in the arousal phase of sexual activity, lack of lubrication in women may be analogous, at least in part, to erectile dysfunction in men. In this study, diminished lubrication was a statistically significant result that remained consistent and independent of age in the analysis; it is also plausible biologically. Thus, we recommend that sexual dysfunction be assessed routinely in hypertensive women by a careful and appropriate sexual history,26 and that it be managed sensitively.

Acknowledgements

We thank the E. Donnall Thomas Resident Research Committee for its support through the protocol development and data collection phase of this project. Ms. Catherine Smith is thanked for her thorough data management and sensitive interviewing skills, as is Ms. Melissa Nichols for her programming support. We also thank Ms. Mary Muller and Mr. Chris Cawley for their considerable effort in typing and manuscript production and Ms. Catherine Mason for expert editorial assistance. Ms. Sara Clark and the medical records staff gave invaluable assistance and support, as did Ms. Cindy St. John of Information Services. Dr. Duncan thanks Dr. D.N. Bateman.

References

1.
Duncan
L
,
Bateman
DN
:
Sexual function in women: do antihypertensive drugs have an impact?
.
Drug Saf
 
1993
;
8
(
3
):
225
234
.
2.
Wakefield
J
:
Sex bias in the diagnosis of primary orgasmic dysfunction
.
Am Psychol
 
1987
;
42
:
464
471
.
3.
Anastos
K
,
Charney
P
,
Charon
RA
,
Cohen
E
,
Jones
CY
,
Marte
C
,
Swiderski
DM
,
Wheat
ME
,
Williams
S
:
Hypertension in women: what is really known? The Women's Caucus, Working Group on Women's Health of the Society of General Internal Medicine
.
Ann Int Med
 
1991
;
115
:
287
293
.
4.
Osborne
M
,
Hawton
K
,
Gath
D
:
Sexual dysfunction among middle aged women in the community
.
Br Med J
 
1988
;
296
:
959
962
.
5.
Laumann
EO
,
Gagnon
JH
,
Michael
RT
,
Michaels
S
The Social Organization of Sexuality: Sexual Practices in the United States
 .
The University of Chicago Press
:
Chicago
,
1994
.
6.
Frank
E
,
Anderson
C
,
Rubinstein
D
:
Frequency of sexual dysfunction in ‘normal’ couples
.
N Engl J Med
 
1978
;
299
(
3
):
111
115
.
7.
Rosen
RC
,
Beck
JG
Patterns of Sexual Arousal
 .
The Guilford Press
:
New York
,
1988
.
8.
Rosen
RC
,
Leiblum
SR
:
Treatment of sexual disorders in the 1990s: an integrated approach
.
J Consult Clin Psych
 
1995
;
63
:
877
890
.
9.
In:
Bancroft
J
(Ed).
Annual Review of Sex Research
 
1991
;
Vol 2
:
Society for the Scientific Study of Sex, Inc
:
Lake Mills, Iowa
.
10.
Beeley
L
:
Drug-induced sexual dysfunction and infertility
.
Adverse Drug React Poisons Rev
 
1984
;
3
:
23
42
.
11.
Smith
PJ
,
Talbert
RL
:
Sexual dysfunction with antihypertensives and antipsychotic agents
.
Clin Pharm
 
1986
;
5
:
373
384
.
12.
Hodge
RH
,
Harward
MP
,
West
MS
,
Krongaard-DeMong
L
,
Kowal-Neeley
MB
:
Sexual function of women taking antihypertensive agents: a comparative study
.
J Gen Int Med
 
1991
;
6
:
290
294
.
13.
Grimm
RH
Jr.
,
Grandits
GA
,
Prineas
RJ
,
McDonald
RH
,
Lewis
CE
,
Flack
JM
,
Yunis
C
,
Svendsen
K
,
Liebson
PR
,
Elmer
PJ
:
Long-term effects on sexual function of five antihypertensive drugs and nutritional hygienic treatment of hypertensive men and women: Treatment of Mild Hypertension Study (TOMHS)
.
Hypertension
 
1997
;
29
: 1:part I:
8
14.
14.
Duncan
LE
,
Lewis
C
,
Smith
CE
,
Jenkins
P
,
Pearson
TA
:
Sex, drugs, and hypertension: a methodological approach for studying a sensitive subject
.
Submitted
 
1999
.
15.
Bulpitt
CJ
,
Dollery
CT
:
Side effects of hypotensive agents evaluated by a self-administered questionnaire
.
Br Med J
 
1973
;
3
:
485
490
.
16.
Croog
SH
,
Levine
S
,
Testa
MA
,
Brown
B
,
Bulpitt
CJ
,
Jenkins
CD
,
Klerman
GL
,
Williams
GH
:
The effects of antihypertensive therapy on the quality of life
.
N Engl J Med
 
1986
;
314
:
1657
1664
.
17.
Leeton
J
,
McMaster
R
,
Worsley
A
:
The effects on sexual response and mood after sterilization of women taking long-term oral contraceptives: results of a double-blind cross-over study
.
Austr NZ J Obstet Gynaecol
 
1978
;
18
:
194
197
.
18.
Bulpitt
CJ
,
Dollery
CT
,
Carne
S
:
A symptom questionnaire for hypertensive patients
.
J Chronic Dis
 
1974
;
27
:
309
323
.
19.
American Psychiatric Association
Diagnostic and Statistical Manual of Mental Disorders (4th ed.)
 .
American Psychiatric Association
:
Washington, DC
,
1994
.
20.
Rosen
RC
,
Taylor
JF
,
Leiblum
SR
,
Bachmann
GA
:
Prevalence of sexual dysfunction in women: results of a survey study of 329 women in an outpatient gynaecological clinic
.
J Sex Marit Ther
 
1993
;
19
:
171
188
.
21.
Taylor
JF
,
Rosen
RC
,
Leiblum
SR
:
Self-report assessment of female sexual function: psychometric evaluation of the brief index of sexual functioning for women
.
Arch Sex Behav
 
1994
;
23
:
627
643
.
22.
Masters
WH
,
Johnson
VE
Human Sexual Inadequacy
 .
Little Brown
:
Boston
,
1970
.
23.
Kaplan
HS
:
Hypoactive sexual desire
.
J Sex Marit Ther
 
1977
;
3
:
3
9
.
24.
Beto
JA
,
Bansul
VK
:
Quality of life in treatment of hypertension: a metaanalysis of clinical trials
.
Am J Hypertens
 
1992
;
5
:
125
133
.
25.
Lewis
C
,
Duncan
LE
,
Ballance
DI
,
Pearson
TA
:
Is sexual dysfunction in hypertensive women uncommon or understudied?
.
Am J Hypertens
 
1998
;
11
:
733
735
.
26.
Prisant
LM
,
Carr
AA
,
Bottini
PB
,
Solursh
DS
,
Solursh
LP
:
Sexual dysfunction with antihypertensive drugs
.
Arch Int Med
 
1994
;
154
:
730
736
.

Author notes

*
This project was funded by an E. Donnall Thomas Research Award, 295-94-T, and by the Research Institute, Bassett Healthcare.