Secondary Logo

Journal Logo

Original Study

Resilience and psychosocial factors linked to symptom experience during the menopause transition for women living with HIV*

King, Elizabeth M. MD1, 2; Kaida, Angela MSc, PhD2, 3; Prior, Jerilynn MD2, 4, 5, 6; Albert, Arianne PhD2; Frank, Peggy BSc, MRM3; Abdul-Noor, Rahma7; Kwaramba, Gladys7; Gormley, Rebecca MPH3, 8; de Pokomandy, Alexandra MD9; Loutfy, Mona MD, MPH10, 11; Murray, Melanie C. M. MD, PhD1, 2, 12

Author Information
doi: 10.1097/GME.0000000000001926

Abstract

The life expectancy of persons living with HIV has dramatically improved; now many women living with HIV (WLWH) are entering menopause (1 year following cessation of menses) and experiencing menopausal symptoms.1 In Canada, approximately one third of WLWH are menopausal, a number that is projected to double in the coming decade.2,3 As these numbers grow, understanding and treating menopausal symptoms will become a priority, particularly when one considers the adverse health impact of negative midlife experiences on quality of life, mood and cognitive performance.4-6 Previous cross-sectional studies suggest that WLWH frequently experience symptoms during menopause,5,7-10 but none have adequately assessed who is most at risk of symptoms, what characteristics are risk factors for intense symptoms, and when these symptoms are likely to occur. These gaps in knowledge leave providers with little understanding of symptom progression and risk factors for severe intensity, both essential to optimize management of symptoms which in turn may benefit holistic health and HIV care.

The menopausal symptom experience is embedded within the social, physical, and emotional environment of each individual; therefore, psychosocial and behavioural evaluation is central to assessing menopausal symptoms for WLWH.9 In literature to date, there has been little attention given to behavioural factors that might positively influence the midlife experience. Resilience is one such behavior, defined as “patterns or processes of positive adaptation…in the context of significant threats to an individual’s life or function.”11 In HIV-negative women, high resilience scores are associated with fewer midlife symptoms,12 however, this link has yet to be assessed in WLWH. In addition to positive influences, factors, such as poverty, violence, substance use, and recent traumatic life events, are also known to relate to severe menopausal symptoms in HIV-negative women.13-15 Although several of these stressors are over-represented in WLWH,16-18 how they modulate symptom experience for this group remains unknown. A study by Miller et al9 begins to assess this by showing that WLWH with history of traumatic life events had more severe menopausal symptoms. However, this study was predominantly with women who used illicit drugs; a similar assessment is warranted for all WLWH, as negative life experiences are common for all, independent of substance use.17 Together, we anticipate that psychosocial and behavioural factors have a strong influence, both protective and harmful, on the experiences and symptoms of midlife WLWH.

In addition to psychosocial and behavioural factors, symptoms are also influenced by midlife reproductive phase; these influences are best considered in conjunction. Perimenopausal and menopausal symptoms change dynamically over years as ovarian hormone levels fluctuate and then decline.19,20 Midlife reproductive phases, defined by expert groups according to menstrual patterns, tend to reflect these hormonal changes; several symptoms have been shown to progress predictably through these phases in HIV-negative women.21-23 Unfortunately, the majority of current evaluations of peri-/menopausal symptoms in WLWH neglect to fully integrate reproductive phases into the assessment.8,9,24 This leads to a fundamental knowledge gap since the progression of menopausal symptoms for WLWH remains largely unknown. Such a description, alongside an assessment of psychosocial factors influencing severity, will provide the first natural history description of peri-/menopause experiences in WLWH.

Taken together, a comprehensive evaluation of influences of midlife symptom severity needs to incorporate psychosocial/behavioural factors and reproductive phase; such an evaluation is currently lacking for WLWH. Furthermore, the progression of menopausal symptoms across phases has not yet been described in this group. To address these central gaps, we conducted a cross-sectional assessment of the midlife experience in WLWH to: i) determine psychosocial and behavioral correlates to symptom severity, adjusted for reproductive phase and ii) elucidate progression of midlife symptoms across reproductive phases.

METHODS

Study participants

The Canadian HIV Women’s Sexual and Reproductive Health Cohort Study (CHIWOS) is a community-based prospective cohort study of WLWH undertaken in three Canadian provinces (British Columbia, Ontario and Quebec) from 2013 to 2018.25 A purposive sampling strategy was used as previously described to ensure recruitment reflected geographic distribution of WLWH across these provinces and included representation of vulnerable and underserved populations.26 Questionnaires were administered by Peer Research Associates at three longitudinal time points or “waves,” timed 18 months apart, of which this analysis uses data from the first two waves. Questionnaires included data on clinical HIV-related parameters, demographics, socioeconomic status, substance use, reported trauma, and other social and structural health determinants.

Participants were included in this analysis if they were assigned female sex at birth and were aged ≥40 years at either of the two waves. The lower age bound of 40 was selected to include women with early menopause yet reduce inaccurate inclusion of younger women with amenorrhea, given its higher prevalence in WLWH compared to the general population.27 As reproductive phase was based around menstrual definitions, women were excluded if they had abnormal menstrual patterns for the following reasons: prior hysterectomy (without oopherectomy), prior bilateral oophorectomy, current pregnancy, current breastfeeding, or hormonal contraception that alters menstrual patterns (ie, progestin-eluting intrauterine device and depo-medroxyprogesterone). History of menopausal hormone therapy (MHT) use was only available for wave two of the study. Given that few WLWH were currently using MHT (18/347) or combined hormonal contraceptives (CHCs, 5/457), these were included in this assessment.

Measures

Perimenopausal and menopausal symptom

The outcome of interest was midlife experiences assessed as a composite measure and by each respective symptom. The composite score was comprised of seven of 11 different symptoms evaluated by the Menopause Rating Scale (MRS), a standardized assessment of peri-/menopausal symptoms validated in WLWH.8,28 The “modified MRS” (mMRS) we used in both wave one and two included the following seven symptoms: hot flashes (including hot flushes and night sweats), cardiac issues, joint and muscle aches, vaginal dryness, bladder problems, depressed mood, and irritability (Supplemental Table 1, https://links.lww.com/MENO/A877).28 Experience of each symptom over the past two weeks was scored as 0 (none), 1 (mild), 2 (moderate), or 3 (severe/very severe), for a summed total symptom score ranging from 0 to 21. For wave two, the standard MRS (scored from 0 to 44) was also calculated to allow comparison with our mMRS for which we used the Pearson correlation test; we also wished to facilitate comparison with previously published cohorts.7,29 Finally, each symptom category was also individually considered.

Midlife reproductive phase classification

Menopause was defined as beginning 1 year after the last or final menstrual flow, in keeping with the World Health Organization definition.30 Women were placed into four categories of the midlife reproductive transition based on final menstrual period (FMP) and self-reported experience changes, adapted from reproductive phases developed by the Stages of Reproductive Aging Workshop (STRAW 10+) expert group.22 These categories were as follows: i) perimenopause (≥40 years old, cycles more than a month apart but menstruated within the last year), ii) 1-2 years after FMP, iii) 2-5 years post-FMP, and iv) >5 years post-FMP. Observations in which timing of menopause was not clear were excluded from our reproductive phase assessment. If women reported having had a period in the past month but selfreported as perimenopausal, they were also included in the perimenopause category.

Covariates

Several potential covariates and predictors of peri-/menopausal symptoms were collected by questionnaire self-report. These included sociodemographic factors (age, ethnicity, housing insecurity, relationship status, education, and employment), psychosocial parameters (depression, resilience, alcohol use [at least 2 times/week], current/former smoking, recent recreational drug use [within past 6 months], experience of HIV-related stigma, and recent physical violence [ie, in past 3 months]), sexual and reproductive health (use of MHT, sexual satisfaction, recent sexual violence [ie, in past 3 months]), and aspects of HIV and clinical health by self-report (body mass index [BMI], CD4 count, viral load (detectable [≥50 copies/mL] vs undetectable [<50 copies/mL]), ART adherence/regimen, coinfections).

Resilience was measured by versions of the resilience scale (RS) developed by Wagnild and Young31 in each study wave. A 10-item scale (RS-10; score range: 10-70) was used for wave one31-33 and an extended version consisting of 14-items (RS-14; score range: 14-98) used for wave two,34 both with established validity and reliability amongst diverse populations (Cronbach’s alpha = 0.86 to 0.96).31,34 Nine items overlapped between the scales. The two RSs were each standardized out of 100 to compare scores between waves and evaluated in tertiles (scores <80, 80-93, >93). Recent traumatic events were defined as experiencing sexual or physical violence within the past three months. Depression was measured using the 10-item Centre for Epidemiologic Studies Depression Scale (CESD 10) on a scale from 0 to 30, using a cut-off of ≥10 for “probable depression” based on previous studies.35 Barriers in access to healthcare was based on a previously described scale of 12 items (scored from 12 to 48) with higher scores meaning less care access.36 HIV stigma was measured using the HIV Stigma Scale, with scores ranging from 0 to 100 and higher scores meaning more stigma.37,38 Finally, sexual satisfaction was asked of all women regardless of their sexual activity and was based on the “contentment” section of the Sexual Satisfaction Scale for Women scale with six questions rated on a 5-point Likert scale; total scores ranged from6 to 30 with sexual satisfaction increasing with higher scores.39

Statistical analysis

Descriptive statistics including mean with standard deviation (SD) for continuous variables (median with interquartile range [IQR] for non-normal distributions) and frequencies with proportions for categorical variables were used to characterize baseline variables. Symptom prevalence was calculated as number of women experiencing at least one peri-/menopausal symptom divided by the total number of women included.

To determine factors predictive of peri-/menopausal symptoms (measured by continuous mMRS), linear mixed effects models were constructed and adjusted for repeated measures comparing symptoms with variables known/expected to be associated with symptom experience. Variables were first assessed in a univariable model and then by multivariable regression to assess their adjusted effects. Multivariable model was adjusted for ethnicity, high BMI (ie, BMI>27), and smoking based on previous data suggesting that these factors correlate with symptom experience.13 Correlates assessed in our models were preselected and included: reproductive phase, annual income, recreational drug use (injection and noninjection), HIV stigma, sexual satisfaction, barriers of access to care, recent physical or sexual violence, depression, and resilience. Depression was dichotomized based on CES-D score (≥10 = yes; <10 = no).35 For ease of interpretation, coefficients for scale-measured variables were reported for a 1-SD score increase. We limited to a single variable when two were collinear (ie, recreational drug use and injection drug use) in the final multivariate model. For comparisons of depression with peri-/menopausal symptoms, models were first built including all menopausal symptoms (including “depressed mood,” measured continuously); secondarily, we removed depressed mood from the symptom composite to avoid collinearity.

For the second objective of assessing symptom by midlife reproductive phase, all observations of menopausal symptoms were evaluated across phases using mixed effects linear regression, controlling for repeated measures. Each symptom type was similarly compared across reproductive phases using mixed effects ordinal logistic regression. When a significant difference was noted across phases, pairwise Tukey tests were conducted to identify the timing of the most significant symptom change.

For all analyses, a P value of < 0.05 was considered statistically significant. For correlates assessed, women had the option to answer: “don’t know” or “prefer not to answer.” Any woman with missing data was excluded from respective analyses. All analyses were performed using R version 3.5.3 (Vienna, Austria).

Ethics approval

The study was approved by research ethics boards at Simon Fraser University, University of British Columbia, Women’s College Hospital, and McGill University Health Centre. Study sites with independent research ethics boards obtained their own approval before commencing enrollment. Informed consent was obtained from all participants at enrollment.

RESULTS

Baseline characteristics

Of 1,422 participants in wave one and 1,244 in wave two, a total of 457 peri-/menopausal women met criteria for inclusion in our study. In the 457 participating women, we examined 703 observations (316 and 387 in waves 1 and 2, respectively, see Fig. 1); 246 women participated in both study waves. After excluding observations with unclear menopausal timing (n = 175), there were 109 observations in perimenopause, 104 in years 1-2 post-FMP, 94 in years 2-5, and 221 >5 years post-FMP.

F1
FIG. 1.:
Flow diagram of participant inclusion in study of menopausal symptoms in women living with HIV in Canada.

Baseline characteristics of participants are summarized in Table 1. Across study waves, there was similar representation from three Canadian provinces (28.4% British Columbia, 38.0% Ontario, 33.6% Quebec). Mean age was 54.7 (±6.6) and BMI 23.6 (±6.0). Approximately one-quarter (24.6%) identified as African/Caribbean/Black, 15.4% as Indigenous, 50.8% as White and 9.2% as mixed race/another ethnicity. The majority (61.6%) had an annual household income below $20,000 and more than half (58.9%) were on social assistance. Forty-four percent (44%) were current smokers and 17.5% reported recent recreational drug use, of which the most common type was stimulants. Depressive symptoms were common; more than half (55.3%) of the women met CES-D criteria for probable depression. Self-reported HIV outcomes revealed that 86.4% reported undetectable viral loads (<50 copies/mL), 81.9% reported a CD4 count >200 cells/mm, and 70.2% had excellent ART adherence (≥95%).

TABLE 1. - Demographics of perimenopausal and menopausal women living with HIV in Canada

Wave

Total N = 703

Wave 1 n = 316

Wave 2 n = 387

Age (y); Mean ± SD

54.7 ± 6.6

54.3 ± 6.4

55.2 ± 6.7

40-45

32 (4.6%)

12 (3.8%)

20 (5.2%)

46-50

136 (19.3%)

72 (22.8%)

64 (16.5%)

51-55

244 (31.9%)

102 (32.3%)

122 (31.5%)

>55

311 (44.2%)

130 (41.1%)

181 (46.8)

Province

British Columbia

200 (28.4%)

92 (29.1%)

108 (27.6%)

Ontario

267 (38.0%)

111 (35.1%)

156 (39.8%)

Quebec

236 (33.6%)

113 (35.8%)

123 (31.3%)

BMI (kg/m2)

23.6 ± 6.0

20.7 ± 4.4

26.1 ± 6.1

Ethnicity

Indigenous

108 (15.4%)

48 (15.2%)

60 (15.3%)

African/Caribbean/Black

173 (24.6%)

69 (21.8%)

104 (26.9%)

White

357 (50.8%)

168 (53.2%)

189 (48.8%)

Other race/ethnicitya

65 (9.2%)

31 (9.8%)

34 (8.8%)

Household income

<$20,000

433 (61.6%)

194 (61.4%)

239 (61.8%)

$20,000-$40,000

167 (23.8%)

76 (24.1%)

91 (23.5%)

>$40,000

84 (11.9%)

36 (11.4%)

48 (12.4%)

Unknown/no answer

19 (2.7%)

10 (3.2%)

9 (2.3%)

Education

High school or less

373 (53.1%)

177 (56.0%)

196 (50.6%)

Trade or technical school

51 (7.3%)

23 (7.3%)

28 (7.2%)

Post-secondary education

275 (39.1%)

114 (36.1%)

161 (41.6%)

Other

4 (0.6%)

2 (0.6%)

2 (0.5%)

Smoking

Current

310 (44.1%)

146 (46.2%)

164 (42.4%)

Former

135 (19.2%)

65 (20.6%)

70 (18.1%)

Never

257 (36.6%)

105 (33.2%)

152 (39.3%)

No answer

1 (0.1%)

0 (0%)

1 (0.3%)

Injection drug use (recent)b

Yes

51 (7.3%)

18 (5.7%)

33 (8.5%)

No

642 (91.3%)

289 (91.2%)

353 (91.0%)

Unknown/no answer

11 (1.6%)

9 (2.8%)

2 (0.5%)

Recreational drug use (recent)b

Yes

123 (17.5%)

54 (17.1%)

69 (17.8%)

No

575 (81.8%)

258 (81.6%)

317 (81.7%)

No answer

6 (0.9%)

4 (1.3%)

2 (0.5%)

Type of recreational drug used

Opioid

16 (13.0%)

7/54 (13.0%)

9/69 (13.0%)

Stimulant

71 (57.7%)

35/54 (64.8%)

36/69 (52.2%)

Polysubstance use

34 (27.6%)

12/54 (22.2%)

22/69 (31.9%)

Depressant

2 (1.6%)

0/54 (0%)

2/69 (2.9%)

Recent experience of physical violence (past 3 months)

Yes

35 (5.0%)

14 (4.4%)

21 (5.4%)

No

613 (87.2%)

278 (88.0%)

335 (86.3%)

Unknown/no answer

55 (7.8%)

24 (7.6%)

31 (8.0%)

Recent experience of sexual abuse (past 3 months)

Yes

15 (2.1%)

5 (1.6%)

10 (2.6%)

No

632 (89.9%)

285 (90.2%)

347 (89.7%)

Unknown/no answer

56 (8.0%)

26 (8.2%)

30 (7.8%)

Probable depressionc

Yes

389 (55.3%)

160 (50.6%)

229 (60.3%)

No

296 (42.1%)

145 (45.9%)

151 (39.7%)

Unknown/no answer

18 (2.6%)

11 (3.5%)

7 (1.8%)

Resilience scored

86.4 [75.0 to 96.1]

86.4 [72.7 to 95.5]

88.9 [77.8 to 96.8]

Midlife Reproductive Phase

Perimenopause

109 (15.5%)

57 (18.0%)

52 (13.4%)

Year 1-2 in menopause

104 (14.8%)

40 (12.7%)

64 (16.5%)

Year 2-5 in menopause

94 (13.4%)

56 (17.7%)

38 (9.8%)

>5 y in menopause

221 (31.4%)

96 (30.4%)

125 (32.3%)

No timing of midlife data

175 (24.9%)

175 (24.9%)

108 (27.9%)

Modified (m) MRS Score

6.0 [3.0 to 10.0]

7.0 [3.0 to 11.0]

6.0 [3.0 to 9.0]

Years living with HIV

15.8 ± 7.1

15.4 ± 7.2

16.2 ± 7.1

Current CD4 count

>500 cells/mm3

384 (54.6%)

173 (54.7%)

211 (54.5%)

200-500 cells/mm3

192 (27.3%)

87 (27.5%)

105 (27.1%)

<200 cells/mm3

33 (4.7%)

19 (6.0%)

14 (3.6%)

Unknown/no answer

94 (13.4%)

37 (11.7%)

57 (14.7%)

Viral load

Undetectable

616 (87.6%)

270 (84.9%)

346 (88.3%)

Detectable

50 (7.1%)

29 (9.1%)

21 (5.4%)

Unknown/no answer

37 (5.3%)

17 (5.3%)

20 (5.2%)

ART adherence

> = 95%

507 (72.1%)

213 (67.0%)

294 (75.0%)

80-94%

96 (13.7%)

51 (16.0%)

45 (11.5%)

50-79%

40 (5.7%)

21 (6.6%)

19 (4.8%)

<50%

12 (1.7%)

9 (2.8%)

3 (0.8%)

Unknown/no answer

48 (6.8%)

22 (6.9%)

26 (6.6%)

Class of current ART

PI

197 (28.0%)

104 (32.7%)

93 (24.0%)

NNRTI

213 (30.3%)

113 (35.5%)

100 (25.8%)

Integrase inhibitor

158 (22.5%)

43 (13.5%)

115 (29.7%)

Other/Unknown/no answer

135 (19.2%)

56 (17.6%)

79 (20.4%)

Data are presented as mean ± standard deviation (SD), median [interquartile range], or n (%) unless otherwise specified.
ART, antiretroviral therapy; BMI, body mass index; CES-D, Centre for Epidemiologic Studies Depression Scale; MRS, Menopause Rating Scale; NNRTI, non-nucleoside reverse transcriptase inhibitors; PI, protease inhibitors; SD, standard deviation.
aOther race/ethnicity included: Chinese/Taiwanese (n = 3), Filipino (n = 1), Japanese (n = 1), Latin American (wave 1 n = 3, wave 2 n = 5), mixed (n = 21), South Asian (n = 1), Southeast Asian (n = 1), and Korean (wave 2 n = 1).
bSelf-reported as drug use in the past 3 months (wave 1) or 6 months (wave 2).
cBased on score of ≥ 10/30 on CES-D scale shown in previous studies to correlate with significant depressive symptoms.35
dResilience score was based on two versions of Wagnild et al resilience score (see methods) and standardized out of 100.31-34

Overall symptom prevalence

Amongst 457 WLWH with records at least once throughout the midlife reproductive transition, 57 (12.5%) reported no symptoms, 400 (87.7%) had one or more symptoms of at least mild severity, 346 (75.7%) of moderate severity, and 250 (54.7%) experienced one or more severe symptoms. The most frequently reported symptoms of mild or higher intensity were joint and muscle aches (reported by 67% in perimenopause), depressed mood (67%), and hot flashes (57%; Table 2). The median mMRS score in both waves was 6 (IQR 3 to 10) on a scale from 0 to 21. Amongst women in wave two, median (standard) MRS was 10 (IQR 5 to 16) on a scale from 0 to 44. Pearson’s correlation test showed a strong correlation between the mMRS and standard MRS (r = 0.96, P<0.0001). Despite approximately half (50.4%) of the women reporting moderate to severe hot flashes during perimenopause, only 21% (81/387) reported ever taking any treatment for these. Eighteen of 387 (4.7%) in wave 2 reported current or recent use of MHT and 30/387 (7.8%) reported ever having taken MHT. Other therapies used for menopausal symptoms were antidepressants (21/387), clonidine (6/387), gabapentin (11/387), and natural health products (21/387). Only 5/457 (1.1%) of women in our cohort reported taking CHC.

TABLE 2. - Percent of women experiencing symptoms of at least mild severity in various midlife reproductive phases and those without symptoms in all phases (n = 345 observations)

Asymptomatic (n = 345)

Perimenopause (n = 94)

1-2 y post-FMP (n = 68)

2-5 y post-FMP (n = 63)

>5 y post-FMP (n = 120)

P valuea

Hot flashes

53.6%

57.4%

50.0%

49.2%

34.2%

<0.0001

Irritability

42.6%

67.0%

52.9%

58.7%

51.7%

0.01

Cardiac complaints

63.2%

43.6%

29.4%

41.3%

33.3%

0.15

Joint and muscle stiffness

33.3%

66.7%

62.7%

68.3%

67.5%

0.36

Vaginal dryness

67.8%

21.1%

26.6%

33.3%

32.1%

N/Ab

Urinary complaints

63.2%

35.1%

33.8%

44.4%

35.8%

0.53

Depressed mood

40.3%

67.0%

44.1%

59.7%

55.5%

0.34

FMP, final menstrual period.
aP value for difference in symptoms across midlife reproductive phases by unadjusted mixed effect regression models.
bSymptom experiences too infrequent to calculate differences between phases.

Correlates of severe symptoms

In the multivariable regression adjusted for midlife reproductive phase, those with higher resilience scores had lower overall symptom experiences (symptom composite 1.37 [2.30 to 0.44] lower; P<0.004; Table 3, Fig. 2). However, those who reported recreational drug use, depressive symptoms, and high sexual satisfaction scores had higher symptom scores (symptom composite 1.71 [0.61 to 2.82] [P = 0.002], 2.89 [2.09 to 3.77] [P<0.001], and 0.46 [0.11 to 0.81] [P = 0.010] higher, respectively; Table 3). Multivariable modeling was repeated with the exclusion of depressed mood from the mMRS (but maintaining CES-D) and showed a consistent, strong association between depressive symptoms, and mMRS (P<0.001). Recent physical or sexual violence was associated with more intense symptoms by univariable analysis but was no longer significantly associated in adjusted analysis (P = 0.218). However, a low rate of recent traumatic events (n = 19) limited assessment. Like resilience, the combination of midlife reproductive phase, depression, and recreational drug use had clear interrelated effects on the midlife symptom experiences of WLWH (Figure 3).

TABLE 3. - Multivariable analysis of factors associated with midlife symptom experience measured by modified menopause rating scale (mMRS) in women living with HIV in Canada

Univariable

Multivariable

Variable

Change in mMRS score (CI)

P value

Change in mMRS score (CI)

P value

Ethnicity (ref: White)

African/Caribbean/Black

-2.03 (-3.04 to -1.01)

<0.001

-0.67 (-1.99 to 0.64)

0.315

Indigenous

1.68 (0.49 to 2.86)

0.006

0.98 (-0.26 to 2.21)

0.122

Other race/ethniticya

0.70 (-0.81 to 2.21)

0.361

0.95 (-0.83 to 2.74)

0.295

BMI (>27)

0.97 (0.94 to 1.01)

0.182

0.85 (0 to 1.71)

0.050

Midlife reproductive phase (ref: perimenopause)

Year 1-2 in menopause

-0.68 (1.92 to 0.57)

0.289

0.01 (-1.03 to 1.04)

0.992

Year 2-5 in menopause

0.07 (-1.27 to 1.42)

0.915

-0.11 (-1.23 to 1.00)

0.841

>5 y in menopause

-1.40 (-2.60 to -0.20)

0.022

-1.06 (-2.05 to -0.08)

0.034

Annual income (ref <$20,000)

Income $20,000-$40,000

0.37 (-0.31 to 1.05)

0.286

Not selected

Income > $40,000

-0.71 (-1.76 to 0.34)

0.183

Smoking

-2.76 (-3.66 to -1.85)

<0.001

-1.06 (-2.28 to 0.16)

0.088

Injection drug use

3.19 (1.75 to 4.63)

<0.001

Not selected

Recreational drug use

3.09 (2.11 to 4.08)

<0.001

1.71 (0.61 to 2.82)

0.002

Probable depressionb

3.36 (2.69 to 4.03)

<0.001

2.89 (2.09 to 3.77)

<0.001

HIV stigmac

1.07 (0.70 to 1.44)

<0.001

0.41 (-0.05 to 0.86)

0.079

Resilience d (ref: low resilience)

Medium (scores 80 to 93)

-0.98 (-1.79 to -0.18)

0.017

-0.75 (-1.57 to 0.01)

0.075

High (scores >93)

-2.88 (-3.74 to -2.03)

<0.001

-1.37 (-2.30 to -0.44)

0.004

Sexual satisfactione

0.30 (-0.06 to 0.67)

0.104

0.46 (0.11 to 0.81)

0.010

Barriers in access to healthcaref

0.74 (0.37 to 1.12)

<0.001

0.38 (-0.01 to 0.76)

0.054

Recent physical or sexual violence

4.60 (3.23 to 5.98)

<0.001

0.95 (-0.56 to 2.47)

0.218

Marginal R2/conditional R2 for multivariate model 0.315/0.740.
BMI, body mass index; CI, confidence interval.
Reference groups as follows: ethnicity (White); menopausal phase (perimenopausal); annual income ($10,000 to $20,000), smoking (current smoker), injection drug use (no), probable depression (no), resilience (low resilience [scores < 80]).
aOther race/ethnicity included: Chinese/Taiwanese, Filipino, Japanese, Latin American, mixed, South Asian, Southeast Asian, and Korean. Variables were measured with the following scales: CES-D 10
b,35 HIV stigma scale
c,37,38 Resiliency Scale
d,31-34 sexual satisfaction
e,39 and barriers to access to care
f.36 For all these scales
c,e,f, coefficients refer to 1 standard deviation increase in the score.

F2
FIG. 2.:
Unadjusted influence of resiliencea and midlife reproductive phasesb on symptom severity as measured by the modified Menopause Rating Scale (mMRS)c for women living with HIV. Error bars extend to 95% confidence intervals of the mean. aHigher scores indicate more resilience. bMidlife reproductive phases include perimenopause, years 1-2, 2-5, >5 from final menstrual period. cHigher mMRS scores indicate more intense symptoms.
F3
FIG. 3.:
Adjusteda effects of recreational drug use, depression and midlife reproductive phasesb on symptom severity as measured by the modified Menopause Rating Scale (mMRS)c for women living with HIV. Error bars extend to 95% confidence intervals of the means. aEstimates were adjusted for resilience, HIV stigma, sexual satisfaction, barriers of access to care, recent physical or sexual violence, smoking, BMI, and ethnicity. bMidlife reproductive phases include perimenopause, years 1-2, 2-5, >5 from final menstrual period. cHigher mMRS scores indicate more intense symptoms.

Progression of symptoms across midlife phases

When symptoms were assessed in women cross-sectionally across midlife reproductive phases of all available observations, there was a significant decrease in mMRS, with symptoms being most intense in perimenopause and generally declining the farther into menopause (P = 0.03). This was driven primarily by hot flashes and irritability, as these particular symptoms declined significantly from perimenopause into more years in menopause (P<0.0001 and P = 0.01). Vaginal dryness, depressed mood, and cardiac, urinary, and joint complaints did not vary significantly by midlife phase (Table 2; Supplemental Figure 2, https://links.lww.com/MENO/A877). More than half (57.4%) of perimenopausal WLWH reported at least mild hot flashes; the proportion of women experiencing hot flashes did not dramatically decline until >5 years in menopause (Table 2). The odds of severe symptoms were significantly higher at every other phase versus >5 years post-FMP; most markedly seen in comparing perimenopause to >5 years in menopause. For women in perimenopause, the odds of being symptomatic were 4.66 (95% CI, 2.36-9.21) compared with women >5 years post-FMP.

DISCUSSION

In this cohort of WLWH ages ≥40 years across Canada, women often experienced multiple peri-/menopausal symptoms of moderate or severe intensity, and symptoms strongly associated with their midlife reproductive phase, positive behavioral factors, and negative psychosocial experiences. Women with higher resilience scores (after adjusting for midlife phase) had less intense symptoms; those reporting recreational drug use, barriers in access to healthcare, and probable depression had more frequent and intense midlife reproductive symptoms. Symptoms were maximal in perimenopause, an effect that was largely driven by hot flashes which did not significantly decrease until four or more years into menopause. These data underscore the importance of evaluating the midlife reproductive symptoms of WLWH within the bio-psychosocial context of each woman’s life.

We demonstrate that resilience is protective against menopausal symptom experience for WLWH, and in doing so, identify the first behavioral factor that has a positive influence on symptom experience for this group. Resilience is a pattern of positive adaption in the context of life adversity and is of particular interest for WLWH, many of whom have faced layers of oppression.11 Increasingly, evidence has emerged demonstrating the health benefits of resilience for WLWH, including increased viral suppression, treatment adherence, and improved health-related quality of life.40-43 For midlife reproductive symptoms, resilience may improve symptom experience by enhancing adaptive coping strategies, social relationships, and willingness to seek health care support.12,44 In addition, resilience may modulate women’s attitudes toward perimenopause and menopause, and, in turn, alter their perception of their symptoms.21,45 Our findings add to the emerging benefits of resilience to the physical health and well-being of WLWH and underscore the importance of developing HIV-related healthcare programs and policies that promote resilience.

In addition to the benefit of positive behaviors on symptom experience, our data confirm the influence of negative psychosocial factors on midlife symptoms, a finding that may partially account for the high rates of symptoms observed in WLWH. In our cohort, nearly all symptom types across midlife reproductive phases were higher than those reported in a prospective population-based cohort of HIV-negative women with similar symptom assessment.23 This echoes a trend previously noted in menopausal WLWH toward higher rates of midlife symptoms than in HIV-negative women, although notable direct comparisons between socio demographically similar HIV + and HIV- women are lacking.8,9,46,47 The reasons behind the high rates of menopausal symptoms for WLWH are not fully elucidated, however, these may in part be driven by psychosocial factors. In an observational analysis, Miller et al9 assesses women with high rates of substance use to show that depression and negative life events were more strongly associated with experience of midlife symptoms than was HIV-status (OR 2.54 [1.64-3.96] and 1.63 [1.18-2.27] vs 1.24 [1.02-1.51]). Our data build upon this finding by showing, in a broader cohort of WLWH, the close association of depression and substance use on symptom experience. We saw a similar trend toward heightened symptoms for women who recently experienced trauma; our data had little potential to show a significant association due to a low event rate and the need to adjust for overlapping comorbidities. We hypothesize that such negative influences decrease a woman’s desire/ability to seek support from health care providers, peers and networks to cope with her midlife symptoms. Experiences such as violence, substance use, and depression disproportionately impact WLWH16-18 and are rarely suffered in isolation.41 Therefore, the cumulative effect of these factors when present together has the potential to profoundly alter women’s midlife experiences.

The combined impact of biologic, psychosocial, and behavioral influences on midlife symptoms underscores the importance of future studies looking at integrating pharmacologic and non-pharmacologic approaches into peri-/menopausal care for WLWH. We observed that hot flashes were of moderate or severe intensity in more than half of perimenopausal WLWH and that these did not significantly decrease until >5 years into menopause. Despite this, a minority of women had received pharmacotherapy for their symptoms and reasons for these low treatment rates remain unexplored.48 Furthermore, our findings draw to question whether nonpharmacologic strategies, particularly those that facilitate resilience, may also augment care by addressing underlying psychosocial/behavioural influences on symptom experience. Unfortunately, there are very few peer-reviewed publications evaluating resilience interventions and none are specific to midlife women or WLWH.49 However, many studies highlight aspects closely associated with resilience that might be leveraged for such interventions. Activities that enhance optimism and positive emotions, such as cognitive behavioral therapy and mindfulness, may be helpful and have been proposed as resilience-building interventions.50,51 Engagement in community through establishing social connections and helping others may also foster resilience, particularly for persons experiencing isolation or loneliness.52 Furthermore, physical activity may also play a role as this is a recognized characteristic of resilient individuals.49,53 Specific to WLWH, a previous study examining correlates to resilience identified that receipt of women-centered HIV care was strongly associated with resilience.41 This care model integrates culturallysafe, trauma-informed, and person-centered care and engages each woman as a partner in her care.54 Such programmatic changes at HIV clinics may go far in fostering a sense of resilience during the healthcare experience.

Our finding that high sexual satisfaction was associated with unwanted, midlife/menopausal symptoms was unexpected. We had hypothesized that menopausal symptoms, particularly genitourinary symptoms such as vaginal dryness, would adversely impact sexual satisfaction. Our findings to the contrary might, in part, be explained by age-related differences in sexual activity and satisfaction, as age is an important predictor of sexual activity in WLWH.55 Our models were limited in controlling for age because of its collinearity with peri-/menopausal phase. Although sexual activity does not always equate to sexual satisfaction,55 we hypothesize that in our analysis, for younger women early in the menopause transition and with more midlife symptoms, higher rates of sexual activity may have accounted for the finding of higher sexual satisfaction.

Our findings should be interpreted in the context of their limitations. First, we have no within-woman measure of experiences before they became perimenopausal or menopausal. As not all symptoms that women experience during the midlife reproductive transition can clearly be attributed to the hormonal changes of perimenopause/menopause (ie, muscle/joint stiffness and depression), future studies should be designed to integrate baseline premenopausal symptomatology to better assess changes with perimenopause. In addition, our analysis lacks an HIV-negative group; further studies are needed to directly compare differences in midlife experiences between WLWH and sociodemographically-similar HIVnegative women. Further, the cross-sectional nature of our assessment may not fully capture the complex and dynamic symptoms experienced across the menopause transition. Although we had two time points, a short follow-up time and limited number of participants involved precluded longitudinal assessment in our analysis; future longitudinal evaluation is warranted to shed further light on changes over time in midlife experiences. We were also limited in our definition of perimenopause which included menstrual cycle irregularities and/or participant self-report. Use of self-report is a limitation as women do not always correctly identify their midlife reproductive phase, however, our observation of high symptomatology in this phase is consistent with the literature23 and argues that many of the women in this phase were indeed perimenopausal. Finally, we included a few women on MHT in our analysis and therefore our symptom reporting may be an underestimate; since low numbers of women used MHT or CHCs, these therapies would have had, at most, a marginal impact on our results.

Despite these limitations, with 703 observations, our assessment provides the largest evaluation of perimenopausal and menopausal symptoms in WLWH to date. The size of the cohort and its purposive sampling strategy make these results highly representative of and generalizable to WLWH in Canada. Further, our comprehensive evaluation of sociodemographic, lifestyle, and health-care related factors contributing to symptom experience provide further guidance to clinical care for symptom screening, counselling, and management in WLWH.

CONCLUSIONS

This large study of WLWH in two waves of assessment 18 months apart shows the intersecting influences of resilience, psychosocial vulnerabilities, and midlife reproductive phase in modulating symptom experiences for midlife WLWH. As the numbers of WLWH entering perimenopause/menopause continue to grow, providers could consider a multifaceted approach to management including pharmacologic and nonpharmacologic strategies, especially those targeted to facilitate resilience. Further research on treatment approaches tailored to the unique midlife reproductive experiences of WLWH are urgently needed and will be an integral part of preserving health-related quality of life for this group.

REFERENCES

1. Eyawo O, Franco-Villalobos C, Hull MW, et al. Changes in mortality rates and causes of death in a population-based cohort of persons living with and without HIV from 1996 to 2012. BMC Infect Dis 2017;17:1–15.
2. Haddad N, Robert A, Weeks A, Popovic N, Siu W, Archibald C. HIV in Canada-surveillance report, 2018. Can Commun Dis Rep 2019;45:304–312.
3. Smit M, Brinkman K, Geerlings S, et al. Future challenges for clinical care of an ageing population infected with HIV: a modelling study. Lancet Infect Dis 2015;15:810–818.
4. Katon JG, Gray KE, Gerber MR, et al. Vasomotor symptoms and quality of life among veteran and non-veteran postmenopausal women. Gerontologist 2016;56:40–53.
5. Maki PM, Rubin LH, Cohen M, et al. Depressive symptoms are increased in the early perimenopausal stage in ethnically diverse human immunodeficiency virus-infected and human immunodeficiency virus-uninfected women. Menopause 2012;19:1215–1223.
6. Rubin LH, Sundermann EE, Cook JA, et al. Investigation of menopausal stage and symptoms on cognition in human immunodeficiency virusinfected women. Menopause 2014;21:997–1006.
7. Duff PK, Money DM, Ogilvie GS, et al. Severe menopausal symptoms associated with reduced adherence to antiretroviral therapy among perimenopausal and menopausal women living with HIV in Metro Vancouver. Menopause 2018;25:531–537.
8. Looby SE, Shifren J, Corless I, et al. Increased hot flash severity and related interference in perimenopausal human immunodeficiency virusinfected women. Menopause 2014;21:403–409.
9. Miller S, Santoro N, Lo Y, et al. Menopause symptoms in HIV-infected and drug-using women. Menopause 2005;12:348–356.
10. Solomon D, Sabin CA, Burns F, et al. The association between severe menopausal symptoms and engagement with HIV care and treatment in women living with HIV. AIDS Care 2021;33:101–108.
11. Masten AS, Wright MO. Resilience over the lifespan: Developmental perspectives on resistance, recovery, and transformation. In: Reich JW, Zautra AJ, Hall JS (editros). Handbook of adult resilience. The Guilford Press; 2010. 213–237.
12. Pérez-López FR, Pérez-Roncero G, Fernández-Iñarrea J, et al. Resilience, depressed mood, and menopausal symptoms in postmenopausal women. Menopause 2014;21:159–164.
13. Gold EB, Sternfeld B, Kelsey JL, et al. Relation of demographic and lifestyle factors to symptoms in a multi-racial/ethnic population of women 40-55 years of age. Am J Epidemiol 2000;152:463–473.
14. Gold EB, Block G, Crawford S, et al. Lifestyle and demographic factors in relation to vasomotor symptoms: baseline results from the study of women’s health across the nation. Am J Epidemiol 2004;159:1189–1199.
15. Pimenta F, Leal I, Maroco J, Ramos C. Menopausal symptoms: do life events predict severity of symptoms in peri- and post-menopause? Maturitas 2012;72:324–331.
16. Zhang Y, Wilson TE, Adedimeji A, et al. The impact of substance use on adherence to antiretroviral therapy among HIV-infected women in the United States. AIDS Behav 2018;22:896–908.
17. Logie CH, Marcus N, Wang Y, et al. A longitudinal study of associations between HIV-related stigma, recent violence and depression among women living with HIV in a Canadian cohort study. J Int AIDS Soc 2019;22:1–12.
18. Fellows RP, Spahr NA, Byrd DA, Mindt MR, Morgello S, Bank MHB. Psychological trauma exposure and co-morbid psychopathologies in HIV+men and women. Psychiatry Res 2015;230:770–776.
19. Prior JC. Perimenopause: the complex endocrinology of the menopausal transition. Endocr Rev 1998;19:397–428.
20. Avis NE, Crawford SL, Greendale G, et al. Duration of menopausal vasomotor symptoms over the menopause transition. JAMA Intern Med 2015;175:531–539.
21. Woods NF, Mitchell ES. Symptoms during the perimenopause: prevalence, severity, trajectory, and significance in women’s lives. Am J Med 2005;118:14–24.
22. Harlow SD, Gass M, Hall JE, et al. Executive summary of the stages of reproductive aging workshop + 10: addressing the unfinished agenda of staging reproductive aging. Menopause 2012;19:387–395.
23. Dennerstein L, Dudley EC, Hopper JL, Guthrie JR, Burger HG. A prospective population-based study of menopausal symptoms. Obstet Gynecol 2000;96:351–358.
24. Johnson TM, Cohen HW, Howard AA, et al. Attribution of menopause symptoms in human immunodeficiency virus-infected or at-risk drugusing women. Menopause 2008;15:551–557.
25. Loutfy M, de Pokomandy A, Kennedy VL, et al. Cohort profile: the Canadian HIV women’s sexual and reproductive health cohort study (CHIWOS). PLoS One 2017;12:1–18.
26. Webster K, Carter A, Proulx-Boucher K, et al. Strategies for recruiting women living with human immunodeficiency virus in community-based research: lessons from Canada. Prog Community Health Partnersh 2018;12:21–34.
27. Cejtin HE, Evans CT, Greenblatt R, et al. Prolonged amenorrhea and resumption of menses in women with HIV. J Womens Health (Larchmt) 2018;27:1441–1448.
28. Schneider HP, Heinemann LA, Rosemeier HP, Potthoff P, Behre HM. The menopause rating scale (MRS): reliability of scores of menopausal complaints. Climacteric 2000;3:59–64.
29. Tuchman E, Pennington LE, Kull RM, Daneshyar S. Relationship between menopause symptoms and HIV risk among midlife women in methadone treatment: a pilot study. Subst Use Misuse 2013;48:711–718.
30. Research on the menopause in the 1990 s. report of a WHO scientific group. World Health Organ Tech Rep Ser 1996;866:1–107.
31. Wagnild GM, Young HM. Development and psychometric evaluation of the resilience scale. J Nurs Meas 1993;1:165–178.
32. Wagnild G. A review of the resilience scale. J Nurs Meas 2009;17:105–113.
33. Kteily-Hawa R, Warren L, Kazemi M, et al. Examining multilevel factors associated with the process of resilience among women living with HIV in a large Canadian cohort study: a structural equation modeling approach. J Int Assoc Provid AIDS Care 2019;18:1–16.
34. Wagnild, Gail M. The eesilience scale user’s guide: for the US english version of the resilience scale and the 14-item resilience scale (RS-14). In: Guinn PE (editor). Resilience center, 2011. Available at: https://www.resiliencecenter.com/products/publications-including-the-true-resilience-book/resilience-scale-users-guide/. Accessed November 23, 2021.
35. Zhang W, O’Brien N, Forrest JI, et al. Validating a shortened depression scale (10 item CES-D) among HIV-positive people in British Columbia, Canada. PLoS One 2012;7:1–5.
36. Heckman TG, Somlai AM, Peters J, et al. Barriers to care among persons living with HIV/AIDS in urban and rural areas. AIDS Care 1998;10:365–375.
37. Wright K, Naar-King S, Lam P, Templin T, Frey M. Stigma scale revised: reliability and validity of a brief measure of stigma for HIV+ youth. J Adolesc Health 2007;40:96–98.
38. Palmer A, Duncan K, Ayalew B, et al. “The way I see it”: the effect of stigma and depression on self-perceived body image among HIV-positive individuals on treatment in British Columbia, Canada. AIDS Care 2021;23:1456–1466.
39. Meston C, Trapnell P. Development and validation of a five-factor sexual satisfaction and distress scale for women: the sexual satisfaction scale for women (SSS-W). J Sex Med 2005;2:66–81.
40. Dale S, Cohen M, Weber K, Cruise R, Kelso G, Brody L. Abuse and resilience in relation to HAART medication adherence and HIV viral load among women with HIV in the United States. AIDS Patient Care STDS 2014;28:136–143.
41. Logie CH, Wang Y, Kazemi M, et al. Exploring social ecological pathways from resilience to quality of life among women living with HIV in Canada. AIDS Care 2018;30:S67–S75.
42. Fumaz CR, Ayestaran A, Perez-Alvarez N, et al. Resilience, ageing, and quality of life in long-term diagnosed HIV-infected patients. AIDS Care 2015;27:1396–1403.
43. McGowan JA, Brown J, Lampe FC, Lipman M, Smith C, Rodger A. Resilience and physical and mental well-being in adults with and without HIV. AIDS Behav 2018;22:1688–1698.
44. Subramaniam S, Camacho LM, Carolan MT, López-Zerón G. Resilience in low-income African American women living and aging with HIV. J Women Aging 2017;29:543–550.
45. Duffy OK, Iversen L, Aucott L, Hannaford PC. Factors associated with resilience or vulnerability to hot flushes and night sweats during the menopausal transition. Menopause 2013;20:383–392.
46. Fantry LE, Zhan M, Taylor GH, Sill AM, Flaws JA. Age of menopause and menopausal symptoms in HIV-infected women. AIDS Patient Care STDS 2005;19:703–711.
47. Ferreira CE, Pinto-Neto AM, Conde DM, Costa-Paiva L, Morais SS, Magalhães J. Menopause symptoms in women infected with HIV: prevalence and associated factors. Gynecol Endocrinol 2007;23:198–205.
48. Howells P, Modarres M, Samuel M, Taylor C, Hamoda H. Experience of hormone replacement therapy in postmenopausal women living with HIV. Post Reprod Health 2019;25:80–85.
49. MacLeod S, Musich S, Hawkins K, Alsgaard K, Wicker ER. The impact of resilience among older adults. Geriatr Nurs 2016;37:266–272.
50. Mancini AD, Bonanno GA. Resilience in the face of potential trauma: clinical practices and illustrations. J Clin Psychol 2006;62:971–985.
51. Tugade MM, Fredrickson BL. Resilient individuals use positive emotions to bounce back from negative emotional experiences. J Pers Soc Psychol 2004;86:320–333.
52. Lutha SS, Cicchetti D. The construct of resilience: implications for interventions and social policies. Dev Psychopathol 2000;12:857–885.
53. Resnick BA, Inguito PL. The resilience scale: psychometric properties and clinical applicability in older adults. Arch Psychiatr Nurs 2011;25:11–20.
54. Women-centred HIV Care. (February 2020). Women-centred HIV care: information for women. Ontario, Toronto. Available at: https://cep.health/clinical-products/hiv/. Accessed June 21, 2021.
55. Kaida A, Carter A, de Pokomandy A, et al. Sexual inactivity and sexual satisfaction among women living with HIV in Canada in the context of growing social, legal and public health surveillance. J Int AIDS Soc 2015;18 (Suppl 5):1–10.
Keywords:

HIV; Perimenopause/menopause; Resilience; Symptoms; Women’s health.

Supplemental Digital Content

© 2022 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The North American Menopause Society.