Menopause Diagnosis: How Doctors Confirm It and What to Do Next

Clinical medical image for womens hrt: Menopause Diagnosis: How Doctors Confirm It and What to Do Next

At a glance

  • Diagnostic standard / 12 months of amenorrhea, age 45 or older, no other cause
  • Average age at menopause / 51.4 years in the United States
  • Perimenopause duration / typically 4 to 8 years before final menstrual period
  • FSH threshold (supportive) / greater than 30 IU/L on two readings 4 to 6 weeks apart
  • Estradiol threshold (supportive) / below 30 pg/mL in the late menopausal transition
  • Hot flashes prevalence / approximately 80% of women during the transition
  • GSM prevalence / affects more than 50% of postmenopausal women
  • First-line vasomotor treatment / systemic estrogen-progestogen therapy (EPT)
  • Insomnia prevalence in menopause / 39 to 47% report chronic sleep disruption
  • Key differential to exclude / thyroid dysfunction (TSH, free T4)

What Actually Defines Menopause Clinically

Menopause has one formal definition: 12 consecutive months without menstruation, occurring naturally at a median age of 51.4 years in US women, once all other causes of amenorrhea have been excluded. No blood test is required to make this call in a woman aged 45 or older with the expected symptom picture.

The STRAW+10 staging system, published in Menopause in 2012 and endorsed by the American Society for Reproductive Medicine, divides the female reproductive lifespan into 10 stages based on cycle variability, hormonal criteria, and symptom patterns [1]. Stages -2 and -1 define the early and late menopausal transition (perimenopause), stage +1a marks the 12-month point confirming menopause, and stage +1b covers the following two years when vasomotor symptoms are typically most severe [1]. Clinicians using STRAW+10 can stage a patient's transition with a menstrual calendar and two hormone values, without relying on symptom severity alone.

For women younger than 45 who stop menstruating, the differential is broader: pregnancy, hypothyroidism, hyperprolactinemia, hypothalamic amenorrhea, premature ovarian insufficiency (POI), and functional hypothalamic suppression from low body weight or high athletic stress must each be considered before a menopause label is applied [2]. POI, defined as ovarian failure before age 40, affects roughly 1 in 100 women and requires FSH confirmation on two occasions at least 4 weeks apart [2].

Which Blood Tests Actually Help

Routine FSH testing is not recommended for women aged 45 and older with classic symptoms and absent or irregular periods. The Menopause Society (formerly NAMS) 2023 position statement states explicitly that laboratory testing is not required for diagnosis in this population [3].

Testing becomes appropriate in three situations: age below 45, an intact uterus with unclear bleeding patterns, and cases where symptoms might reflect thyroid disease rather than ovarian aging. A serum FSH above 30 IU/L drawn in the follicular phase of a remaining cycle, confirmed on a second draw 4 to 6 weeks later, supports the diagnosis of the late transition or menopause [3]. Estradiol below 30 pg/mL on the same draw adds specificity. Because FSH fluctuates widely during perimenopause, a single normal FSH does not exclude the diagnosis.

TSH and free T4 belong in the initial workup for any woman presenting with hot flashes, fatigue, weight change, or mood disturbance. Hypothyroidism and hyperthyroidism can each produce sweating episodes, palpitations, sleep disruption, and mood changes that are clinically indistinguishable from vasomotor symptoms without a blood test [4]. The American Thyroid Association recommends TSH screening for women over 35 every five years, and more frequently when symptoms arise [4].

Anti-Müllerian hormone (AMH) levels decline years before FSH rises, but AMH is not yet validated as a standalone diagnostic criterion for menopause staging in clinical practice [5]. It may be useful in fertility counseling for women in their late 30s and early 40s, but it does not replace FSH-based criteria.

Perimenopause: Symptoms That Precede the Diagnosis by Years

Perimenopause begins with cycle irregularity, not with the cessation of periods. Most women notice their first changes in their mid-to-late 40s, though the transition can start as early as 40 in some cases. The Study of Women's Health Across the Nation (SWAN), a longitudinal cohort of 3,302 women followed for over two decades, documented that the late transition (cycles varying by 60 days or more) lasted a median of 3.8 years before the final menstrual period [6].

Characteristic early perimenopause symptoms include:

  • Cycles shortening by 2 to 7 days, then becoming unpredictable
  • New-onset premenstrual syndrome or worsening of existing PMS
  • Sleep fragmentation, particularly early-morning waking
  • Mood lability disproportionate to life stressors
  • Brain fog and word-retrieval difficulty
  • Breast tenderness from fluctuating estrogen surges

These symptoms arise before estrogen actually falls. In early perimenopause, estrogen levels can be paradoxically elevated due to compensatory FSH stimulation of remaining follicles [6]. The erratic cycling of estrogen, not a simple linear decline, drives most early symptoms. Women who present in this phase often receive incorrect reassurances that "hormones look normal" because a single mid-cycle draw may show normal or elevated estradiol.

The late transition (STRAW+10 stage -1) begins when a woman misses a period by 60 days or more. Vasomotor symptoms, genitourinary changes, and sleep disruption become more prominent at this stage [1]. Mean FSH rises above 25 IU/L and estradiol begins its sustained decline.

Hot Flashes: Mechanism, Severity, and Treatment Options

Hot flashes affect approximately 80% of women during the menopause transition and persist for a median of 7.4 years from onset, according to the SWAN Daily Hormone Study [7]. For about 25% of women, episodes are severe enough to interfere with work, relationships, and quality of life. Duration is longer in women who start experiencing flashes before their final period [7].

The mechanism centers on thermoregulatory dysfunction in the hypothalamus. Declining estrogen narrows the thermoneutral zone, meaning a smaller rise in core body temperature triggers a heat-dissipation response: peripheral vasodilation, sweating, and a subjective sensation of intense warmth spreading from the chest upward [8]. The kisspeptin/neurokinin B/dynorphin (KNDy) neuron population in the hypothalamic infundibular nucleus appears to be the primary site where estrogen withdrawal drives this dysregulation [8].

Systemic hormone therapy remains the most effective pharmacologic treatment. In the KEEPS trial (Kronos Early Estrogen Prevention Study, N=727), oral conjugated equine estrogen 0.45 mg/day and transdermal estradiol 50 mcg/day each reduced hot flash frequency by more than 70% compared to placebo over 48 months [9]. The Menopause Society states: "Hormone therapy is the most effective treatment for vasomotor symptoms and is appropriate for healthy symptomatic women under age 60 or within 10 years of menopause onset" [3].

Fezolinetant (Veozah), a neurokinin-3 receptor antagonist, received FDA approval in May 2023 as the first non-hormonal prescription treatment targeting the KNDy pathway directly [10]. In the SKYLIGHT 1 and SKYLIGHT 2 trials (combined N=1,022), fezolinetant 45 mg daily reduced moderate-to-severe hot flash frequency by 59% versus 40% for placebo at week 12 [10]. It offers an option for women with hormone-sensitive cancers or those who decline hormone therapy.

Venlafaxine 37.5 to 75 mg daily and paroxetine 7.5 mg daily (FDA-approved specifically for vasomotor symptoms as Brisdelle) reduce hot flash frequency by 40 to 60% in randomized trials, making them the preferred non-hormonal options when fezolinetant is unavailable [11]. Gabapentin 300 mg three times daily provides similar efficacy in some patients, particularly when insomnia is the dominant complaint [11].

Behavioral measures that reduce trigger exposure, including limiting caffeine after noon, avoiding alcohol within three hours of sleep, and keeping the bedroom below 67°F (19.4°C), can reduce frequency by 10 to 20% and should accompany any pharmacologic plan [12].

Genitourinary Syndrome of Menopause (GSM): Diagnosis and Management

GSM replaces the older terms "atrophic vaginitis" and "vulvovaginal atrophy" because it more accurately reflects the full scope of estrogen-deficiency changes affecting the vulva, vagina, bladder, and urethra. More than 50% of postmenopausal women develop clinically significant GSM, and unlike vasomotor symptoms, it does not improve spontaneously over time [13].

Diagnostic criteria are clinical: thinning and pallor of vaginal epithelium, reduced rugation, petechiae, a vaginal pH above 5.0, and symptoms including dryness, burning, discharge, dyspareunia, urinary urgency, or recurrent UTIs [13]. A vaginal pH strip and a brief pelvic exam are sufficient for diagnosis in most outpatient settings. Vaginal cytology showing a reduced maturation index (fewer superficial cells) can confirm atrophy but is not required before starting treatment.

Local vaginal estrogen is first-line therapy and carries a strong safety profile even in breast cancer survivors when systemic absorption is minimal. Options include:

  • Estradiol vaginal cream 0.01% (Estrace), 0.5 g inserted 2 to 3 times weekly after a loading phase
  • Estradiol vaginal tablet 10 mcg (Vagifem/Yuvafem), inserted twice weekly
  • Estradiol vaginal ring 7.5 mcg/day (Estring), replaced every 90 days
  • Prasterone (Intrarosa) vaginal insert 6.5 mg, inserted nightly; an FDA-approved DHEA formulation that converts locally to estrogen and androgen [14]
  • Ospemifene (Osphena) 60 mg oral daily, a selective estrogen receptor modulator approved for moderate-to-severe dyspareunia from GSM [14]

A 2018 Cochrane review of 30 trials (N=6,235) found that local vaginal estrogen preparations produced statistically equivalent symptom relief compared to each other, with all formulations significantly outperforming placebo on vaginal dryness, dyspareunia, and pH normalization (P<0.001) [15]. The Menopause Society 2020 position statement on GSM notes that local estrogen does not require a progestogen for endometrial protection at standard doses [13].

Non-prescription options, specifically vaginal moisturizers containing hyaluronic acid or polycarbophil used three times weekly, and silicone-based lubricants used at the time of intercourse, reduce symptom burden and may be combined with prescription therapy [15].

Menopause-Related Insomnia: Why It Happens and How to Treat It

Between 39% and 47% of perimenopausal and postmenopausal women report chronic sleep disruption, compared to 33% of premenopausal women in the same SWAN cohort analyses [16]. Sleep architecture changes with menopause independently of nocturnal hot flashes, meaning some women with minimal vasomotor symptoms still develop significant insomnia [16].

Contributing mechanisms include:

  1. Hot-flash-driven arousals, which fragment slow-wave and REM sleep
  2. Declining progesterone, which has direct GABAergic sleep-promoting effects
  3. Mood disorders, particularly anxiety and depression, which become more common during the transition
  4. Primary sleep disorders like restless legs syndrome (prevalence doubles in perimenopause) and obstructive sleep apnea (risk rises after menopause due to upper-airway tissue changes)

Cognitive behavioral therapy for insomnia (CBT-I) is the first-line treatment for chronic insomnia regardless of menopausal status, recommended over pharmacotherapy by the American College of Physicians [17]. A 2019 meta-analysis of 87 randomized trials (N=6,873) confirmed CBT-I produced a mean reduction in sleep onset latency of 19.0 minutes and improved sleep efficiency by 9.9 percentage points, with effects maintained at 12-month follow-up [17].

Systemic hormone therapy addresses the underlying hormonal driver and reduces nocturnal awakenings independently of its effect on hot flashes. The Women's Health Initiative Insomnia Rating Scale data showed that women assigned to EPT reported significantly fewer sleep disturbance symptoms at one year compared to placebo, with 32% reporting improvement versus 21% in the placebo arm [18].

Melatonin receptor agonists (ramelteon 8 mg) and low-dose doxepin 3 to 6 mg have FDA approval for insomnia and carry lower dependency risk than benzodiazepines, making them reasonable short-term adjuncts [19]. Benzodiazepines and Z-drugs should be used cautiously in women over 55 given fall risk and rebound insomnia on discontinuation [19].

Sleep hygiene measures with specific evidence in menopausal women include maintaining a consistent wake time (even on weekends), restricting time in bed to actual sleep time plus 30 minutes in the first two weeks of CBT-I, and cooling the bedroom to below 67°F [12].

Mood Changes, Cognitive Symptoms, and What They Indicate

Depression risk approximately doubles during the menopausal transition compared to premenopausal years, independent of prior depressive history. The Harvard Study of Moods and Cycles (N=460) found that women without prior depression had 1.8 times the odds of developing a major depressive episode during perimenopause compared to premenopausal women of the same age [20].

Fluctuating estradiol, not simply low estradiol, drives much of the mood vulnerability. Estradiol modulates serotonin transporter density and MAO-A activity in limbic circuits; rapid drops in estradiol may precipitate dysphoria in susceptible individuals [20]. Women with a history of premenstrual dysphoric disorder (PMDD) or postpartum depression have heightened sensitivity to hormonal fluctuations and are at particularly elevated risk.

Cognitive complaints, specifically word retrieval difficulty, working memory lapses, and reduced processing speed, are reported by 44 to 62% of women during perimenopause in SWAN cognitive substudy data [21]. These symptoms are most pronounced in the late transition and early postmenopause and tend to improve modestly as the brain adapts to lower estrogen levels over 2 to 4 years [21]. Persistent severe cognitive symptoms warrant formal neuropsychological screening to exclude early neurodegenerative disease.

Estrogen therapy started during the menopause transition (the "critical window hypothesis") may preserve cognitive function, but the evidence does not yet support starting hormone therapy solely for dementia prevention. The 2022 Menopause Society consensus statement advises against initiating HRT in women over 65 primarily for cognitive neuroprotection [3].

Hormone Therapy: Who Qualifies and Which Regimen to Choose

Candidacy for systemic hormone therapy involves weighing symptom severity, time since menopause, cardiovascular risk, breast cancer risk, and uterine status. The decision framework used at HealthRX organizes this as follows:

Women with a uterus require combined estrogen-progestogen therapy (EPT) to prevent endometrial hyperplasia. Continuous combined EPT (estradiol plus micronized progesterone 100 mg or a synthetic progestin daily) is the standard regimen and achieves endometrial protection within 12 months [3]. The PEPI trial (N=875) confirmed that unopposed estrogen raised endometrial hyperplasia rates to 62% over three years versus 1% with combined therapy [22].

Women without a uterus (post-hysterectomy) may use estrogen-only therapy (ET), which avoids progestogen-related side effects and carries a more favorable breast safety profile based on the Women's Health Initiative ET arm data [18].

Route of administration affects cardiovascular and VTE risk meaningfully. Transdermal estradiol (patches, gels, sprays) bypasses hepatic first-pass metabolism, does not raise CRP or triglycerides, and in observational data from the E3N cohort (N=80,377), was not associated with increased VTE risk, while oral estrogen was [23]. The Menopause Society recommends transdermal delivery for women with elevated cardiovascular risk, obesity (BMI >30), a personal history of migraine with aura, or any prior VTE [3].

Micronized progesterone (Prometrium, Utrogestan) appears to carry a lower breast cancer risk than synthetic progestins based on data from the Million Women Study and the E3N cohort, though no head-to-head randomized trial has directly confirmed this [23]. The absolute annual risk increase for breast cancer with combined EPT is approximately 0.8 per 1,000 women per year based on WHI data, comparable to the risk associated with drinking one glass of wine daily [18].

Testosterone in women remains off-label in the United States but is supported by the Global Consensus Position Statement on Testosterone Use in Women (2019), which concluded that transdermal testosterone at physiologic doses (targeting a serum level of 5 to 25 ng/dL) improves sexual desire, arousal, and orgasm frequency in postmenopausal women with hypoactive sexual desire disorder [24].

Differentiating Menopause from Thyroid Disease

Hypothyroidism and hyperthyroidism each produce symptom clusters that overlap substantially with the menopause transition. Hypothyroidism causes fatigue, weight gain, mood depression, dry skin, and cold intolerance. Hyperthyroidism produces sweating, palpitations, anxiety, heat intolerance, and sleep disturbance. Both conditions can cause menstrual irregularity that mimics the pattern of perimenopause.

The American Thyroid Association prevalence data place subclinical hypothyroidism at 4 to 8.5% in women over 40, with overt hypothyroidism at approximately 1 to 2% [4]. These rates are high enough that thyroid disease represents a real and common differential rather than an academic one.

A serum TSH is the single best screening test. A normal TSH (0.4 to 4.0 mIU/L) with no other clinical red flags effectively excludes thyroid disease as the primary driver of symptoms, though some clinicians check free T4 when TSH sits at the high or low end of the reference range. Women should not receive a menopause diagnosis by exclusion alone without TSH testing.

Autoimmune thyroiditis (Hashimoto disease) deserves specific mention because it peaks in incidence in women aged 30 to 50 and can produce fluctuating thyroid function that creates a confusing clinical picture during perimenopause. Checking thyroid peroxidase antibodies (TPOAb) in women with a family history of autoimmune disease, a goiter, or TSH values trending upward on serial measurements adds diagnostic clarity [4].

Frequently asked questions

How is menopause officially diagnosed?
Menopause is diagnosed clinically after 12 consecutive months without a menstrual period in a woman aged 45 or older, with no other medical explanation for the amenorrhea. Blood tests are not required for women in this age group with expected symptoms but are used in ambiguous cases or when the woman is younger than 45.
What FSH level indicates menopause?
An FSH level above 30 IU/L on two separate draws taken 4 to 6 weeks apart, combined with an estradiol below 30 pg/mL, supports a diagnosis of late menopausal transition or confirmed menopause. Because FSH fluctuates widely during perimenopause, a single normal reading does not exclude the diagnosis.
What are the first signs of perimenopause?
The earliest signs are usually cycle changes: periods arriving 2 to 7 days earlier than usual, then becoming unpredictable. New or worsening premenstrual syndrome, sleep fragmentation, mood lability, and early-morning waking often precede hot flashes by one to three years.
Can you have menopause symptoms with normal hormone levels?
Yes. In early perimenopause, estradiol may be normal or even elevated due to compensatory FSH stimulation, yet symptoms like insomnia, mood changes, and cycle irregularity are already present. A single hormone panel does not rule out the menopausal transition in a woman with characteristic symptoms.
What is the best treatment for menopausal hot flashes?
Systemic hormone therapy (estrogen with progestogen in women with a uterus) is the most effective option, reducing hot flash frequency by more than 70% in clinical trials. For women who cannot use hormones, fezolinetant 45 mg daily (FDA-approved 2023), venlafaxine 37.5 to 75 mg daily, or paroxetine 7.5 mg daily are evidence-supported alternatives.
What is genitourinary syndrome of menopause (GSM)?
GSM describes the collection of vulvovaginal and urinary symptoms caused by estrogen deficiency, including vaginal dryness, burning, dyspareunia, urinary urgency, and recurrent UTIs. It affects more than 50% of postmenopausal women and does not resolve without treatment, unlike vasomotor symptoms which may diminish over time.
Is local vaginal estrogen safe for breast cancer survivors?
Most oncology and menopause societies consider low-dose local vaginal estrogen (10 mcg estradiol tablet or 0.5 g cream twice weekly) acceptable for breast cancer survivors with severe GSM, particularly those not on [aromatase inhibitors](/classes-aromatase-inhibitors/class-overview-monograph), because systemic absorption is minimal. Women on aromatase inhibitors should discuss ospemifene or prasterone as alternatives with their oncologist.
Why does menopause cause insomnia?
Menopause-related insomnia has several overlapping causes: nocturnal hot flashes fragment slow-wave and REM sleep, declining progesterone reduces GABAergic sleep promotion, and depression and anxiety become more common during the transition. Primary sleep disorders including restless legs syndrome and obstructive sleep apnea also increase in prevalence after menopause.
Does menopause cause depression?
The menopausal transition approximately doubles the risk of a major depressive episode compared to the premenopausal years, independent of prior depression history. Rapidly fluctuating estradiol, not simply low estradiol, disrupts serotonin and dopamine pathways in limbic circuits. Women with a history of PMDD or postpartum depression are at the highest risk.
How long does perimenopause last?
The SWAN study found that the late menopausal transition lasted a median of 3.8 years before the final menstrual period. When early transition changes are included, total perimenopause duration ranges from 4 to 8 years for most women, though some experience a transition of 10 years or longer.
Can thyroid disease be mistaken for menopause?
Thyroid disease can produce sweating episodes, palpitations, fatigue, weight change, mood disturbance, and menstrual irregularity that closely mimic perimenopause symptoms. A serum TSH is the minimum required test to distinguish the two. Subclinical hypothyroidism affects 4 to 8.5% of women over 40, making this a clinically relevant differential.
Is hormone therapy safe for women in their 50s?
The Menopause Society states that hormone therapy is appropriate for healthy symptomatic women under age 60 or within 10 years of menopause onset and without contraindications. The absolute risk increase for breast cancer with combined EPT is approximately 0.8 per 1,000 women per year, and transdermal estradiol appears to carry no increased VTE risk in observational data.
What blood tests should be done for menopause evaluation?
For women aged 45 or older with classic symptoms, the minimum workup is TSH to exclude thyroid disease. FSH and estradiol are added when the diagnosis is unclear, the woman is under 45, or premature ovarian insufficiency is suspected. A complete metabolic panel and lipid panel are useful for baseline cardiovascular risk assessment before starting hormone therapy.

References

  1. 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(4):387-395. https://pubmed.ncbi.nlm.nih.gov/22343510/
  2. European Society of Human Reproduction and Embryology. Guideline on the management of premature ovarian insufficiency. 2015. https://pubmed.ncbi.nlm.nih.gov/26116891/
  3. The Menopause Society. The 2023 Menopause Society Position Statement on Hormone Therapy. Menopause. 2023;30(6):573-590. https://pubmed.ncbi.nlm.nih.gov/37226408/
  4. Garber JR, Cobin RH, Gharib H, et al. Clinical practice guidelines for hypothyroidism in adults: sponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association. Endocr Pract. 2012;18(Suppl 3):1-207. https://pubmed.ncbi.nlm.nih.gov/23246686/
  5. Dewailly D, Andersen CY, Balen A, et al. The physiology and clinical utility of anti-Müllerian hormone in women. Hum Reprod Update. 2014;20(3):370-385. https://pubmed.ncbi.nlm.nih.gov/24430863/
  6. Harlow SD, Crawford SL, Sommer B, et al. Recommendations from a multi-study evaluation of proposed criteria for staging reproductive aging. Climacteric. 2005;8(1):7-25. https://pubmed.ncbi.nlm.nih.gov/15804727/
  7. Avis NE, Crawford SL, Greendale G, et al. Duration of menopausal vasomotor symptoms over the menopause transition. JAMA Intern Med. 2015;175(4):531-539. https://pubmed.ncbi.nlm.nih.gov/25686030/
  8. Rance NE, Dacks PA, Mittelman-Smith MA, Romanovsky AA, Krajewski-Hall SJ. Modulation of body temperature and LH secretion by hypothalamic KNDy (kisspeptin, neurokinin B and dynorphin) neurons: a novel hypothesis on the mechanism of hot flushes. Front Neuroendocrinol. 2013;34(3):211-227. https://pubmed.ncbi.nlm.nih.gov/23978487/
  9. Harman SM, Black DM, Naftolin F, et al. Arterial imaging outcomes and cardiovascular risk factors in recently menopausal women: a randomized trial. Ann Intern Med. 2014;161(4):249-260. https://pubmed.ncbi.nlm.nih.gov/25069991/
  10. FDA. FDA approves fezolinetant (Veozah) for vasomotor symptoms due to menopause. May 2023. https://www.accessdata.fda.gov/drugsatfda_docs/label/2023/216578s000lbl.pdf
  11. Pinkerton JV, Joffe H, Kazempour K, Mekonnen H, Bhaskar S, Lippman J. Low-dose paroxetine (7.5 mg) improves sleep in women with vasomotor symptoms. Menopause. 2015;22(1):50-58. https://pubmed.ncbi.nlm.nih.gov/25003620/
  12. Ensrud KE, Joffe H, Guthrie KA, et al. Effect of escitalopram on insomnia symptoms and subjective sleep quality in healthy perimenopausal and postmenopausal women with hot flashes. Menopause. 2012;19(8):848-855. https://pubmed.ncbi.nlm.nih.gov/22453200/
  13. Portman DJ, Gass ML; Vulvovaginal Atrophy Terminology Consensus Conference Panel. Genitourinary syndrome of menopause: new terminology for vulvovaginal atrophy from the International Society for the Study of Women's Sexual Health and the North American Menopause Society. Menopause. 2014;21(10):1063-1068. https://pubmed.ncbi.nlm.nih.gov/25160739/
  14. FDA. Prasterone (Intrarosa) prescribing information. 2016. https://www.accessdata.fda.gov/drugsatfda_docs/label/2016/208470s000lbl.pdf
  15. Lethaby A, Ayeleke RO, Roberts H. Local oestrogen for vaginal atrophy in postmenopausal women. Cochrane Database Syst Rev. 2016;(8):CD001500. https://pubmed.ncbi.nlm.nih.gov/27577677/
  16. Kravitz HM, Ganz PA, Bromberger J, Powell LH, Sut