Implications of recent health policies for women’s reproductive mental health

Abstract

Background:

The economic cost of perinatal mood and anxiety disorders (PMADs) is high and includes the cost of reduced maternal economic productivity, more preterm births, and increases in other maternal mental health expenditures. PMADs also substantially contribute the cost of maternal morbidity. This paper offers a discussion of the quality-of-care cascade model of PMADs, which outlines care pathways that people typically face as well as gaps and unmet needs that frequently happen along the way. The model uses the US health system as an example. A discussion of international implications follows.

Discussion:

The quality-of-care cascade model outlines downward dips in quality of care along the perinatal mental health treatment continuum, including access (many Americans do not have access to affordable health insurance), enrollment (even when individuals are offered health insurance, some do not enroll), coverage (even if individuals have health insurance, some needed services or providers may not be covered), choice (even if services and providers are covered, patients may not be able to choose among plans, institutions, or clinicians), consistency (even if patients have a choice of plan or provider, a consistent source of care may not be accessible), referral (even if care is available and accessible, referral services may not be), quality (even if patients have access to both care and referral services, there may be gaps in the quality of care provided), adherence (even if patients receive high-quality care, they may not be adherent to treatment), barriers (societal forces that may influence people’s choices and behaviors), and shocks (unanticipated events that could disrupt care pathways). In describing the quality-of-care cascade model, this paper uses the US healthcare system as the primary example. However, the model can extend to examine quality-of-care dips along the perinatal mental health treatment continuum within the international context. Although the US healthcare system may differ from other healthcare systems in many respects, shared commonalities lead to quality-of-care dips in countries with healthcare systems structured differently than in the US.

Implications for Health Policies:

The global cost of PMADs remains substantial, and addressing the costs of these conditions could have a significant impact on overall cost and quality of care internationally. The quality-of-care cascade model presented in this paper could help identify, understand, and address the complex contributing factors that lead to dips in quality-of-care for perinatal mental health conditions across the world.

The societal burden of perinatal mood and anxiety disorders

Lack of treatment for perinatal mood and anxiety disorders (PMADs) leads to a substantial economic burden.[1, 2] This paper offers a discussion of the quality-of-care cascade model applied to PMADs, which outlines care pathways that people typically face as well as gaps and unmet needs that frequently happen along the way. The model uses the US health system as an example, and a discussion of international implications that includes a review of the substantial global cost of PMADs internationally follows. The quality-of-care cascade model presented in this paper could help identify, understand, and address the complex contributing factors that lead to dips in quality-of-care for perinatal mental health conditions across the world. Each step of the cascade represents a potential target for policymakers to improve quality of care and related cost effectiveness. Previous studies have used similar cascade models in health system economic modeling analysis to identify barriers to care and outline targeted incremental or comprehensive policy approaches[3] as well as quality measurement and quality improvement goals.[4]

In the United States alone, PMADs cost $14 billion nationally for the 2017 birth cohort from conception to 5 years postpartum. The average cost per affected mother-child dyad reached $31,800. The largest costs included reduced economic productivity among affected mothers, more preterm births, and increases in other maternal health expenditures. Several states have assessed costs associated with PMADs. For example, in Texas, PMADs affected 17.2% of mothers enrolled in Texas’ Medicaid for Pregnant Women and cost $962 million between conception and the first five years of a child’s life. [5]

PMADs also substantially contribute to the costs of maternal morbidity.[6] The total cost of nine maternal morbidity conditions including amniotic fluid embolism; cardiac arrest; gestational diabetes; hemorrhage; hypertensive disorders; maternal mental health conditions; renal disease; sepsis; venous thromboembolism for all pregnancies and births in 2019 reached $32.3 billion from conception to five years postpartum, amounting to $8,624 in societal costs per birthing person.

People with mental health conditions use more non-mental health services than people without mental health conditions[7], and individuals with PMADs have higher health care costs than those without those conditions.[1, 8] Costs associated with having a baby, regardless of mental health status, remain high. Some call this financial hardship or toxicity associated with giving birth.[9, 10]

Advocates, providers, and educators can share this data about the significant impact of PMADs with policymakers, health systems, and other stakeholders to encourage addressing unmet needs for mothers, children, and families. The economic case for high-quality affordable and accessible mental health care in the perinatal period and throughout life can encourage policymakers, including state and federal executives and legislators, and health system leadership, to expand access to detection, diagnosis, and treatment for affected individuals.

Conceptual model

Within the American health system, health policies affect treatment and outcomes for delivering individuals. To provide context, this paper describes a model that outlines care pathways that people typically face, and gaps and unmet needs that all too frequently happen along the way. This model uses the US health system as an example, but the insights may also translate globally.

In a landmark piece published in 2000 in the Journal of the American Medical Association titled, “Transforming insurance coverage into quality health care: voltage drops from potential to delivered quality”[11], Dr. John Eisenberg, then Director of the Agency for Healthcare Research and Quality, argues that having access to health insurance does not necessarily lead to receiving high-quality care. Eisenberg contrasts the enormous potential of the US healthcare system with the unacceptable disparities in quality of care received, an issue that remains today over two decades later. His words highlighted the vulnerability of the US healthcare system, where high-quality care can become lost, along with recommendations to recognize and prevent each voltage drop.

Others have changed his concept using the language and imagery of cascades of care rather than electric jolts. Based on studies in the literature, the cascade concept appears to work in multiple ways. One way involves patients receiving too much treatment, such as one medication causing a side effect that leads to the use of another and then another.[12] However, other studies use characterizations of treatment that align more with Eisenberg’s voltage drop model regarding losses in quality of care at each dip downward in the cascade.[13–16]

Another notable article relevant for this discussion uses the cascade metaphor. Cox et al. published an article in 2016 in the Journal of Clinical Psychiatry titled, “The Perinatal Depression Treatment Cascade: Baby Steps Toward Improving Outcomes.”[17] This piece grew out of another related article by the authors on the cascade of depression treatment more broadly,[18] with the former article highlighting the unique challenges faced during the perinatal period. Cox et al. conducted a systematic search of existing articles to highlight issues at each decrement within the cascade, terminology useful when discussing gaps identified between detection, diagnosis, and treatment for PMADs.

This paper presents an adaptation and expansion of downward dips in the quality-of-care cascade along the perinatal mental health treatment continuum. Further, these downward dips occur within both broader social forces, such as racism and stigma, and in relation to external shocks, including policy changes and public health events, such as the ongoing COVID-19 global pandemic. The following sections explore how these dips impact birthing people within the US healthcare system. A discussion of international applications follows.

Step 1. Access: Many Americans do not have access to affordable health insurance

Despite recent gains in the number of people who have access to health insurance, 26 million or 8% of people in the US lack health insurance.[19] The rate includes 5% of children uninsured in 2021.[20] Although this represents an all-time low number of uninsured people, over one in 10 women still do not have health insurance.[21]

Although 41 states including Washington DC have expanded Medicaid as of September 2023, 10 states have not. This leaves millions of individuals without access to affordable health insurance through employers, the health care marketplace, or public insurance options.[22, 23] Therefore, coverage gaps remain for individuals not eligible for Medicaid. Although pregnancy represents a qualifying event for individuals to access Medicaid coverage, it does not guarantee coverage before pregnancy or beyond 60 days after delivery.

The American Rescue Plan Act of 2021 gave states the flexibility to extend Medicaid from 60 days to one year postpartum.[24] As of September 2023, 37 states, including Washington DC, implemented this 12-month extension, with 9 more states planning to implement an extension and 2 adopting a limited extension.[25]

Although access to insurance does not guarantee good health outcomes, not having insurance can lead to a host of challenges and undesired outcomes for mother and child.

Step 2. Enrollment: Even when individuals are offered health insurance, some do not enroll

Even after the Affordable Care Act expanded health insurance options, some people choose not to select a health plan, even when facing a potential tax penalty for not enrolling.

The 2022 Commonwealth Fund biennial health insurance survey asked people why they did not buy marketplace coverage or why they dropped their coverage. Costs represent the primary reason people either did not buy or dropped coverage.[26] Among uninsured individuals or those with a coverage gap who previously had Medicaid, they most often cited lack of eligibility as their reason for lack of coverage.

Administrative burdens, such as complex and time-consuming paperwork can effectively prevent enrollment in insurance plans,[27, 28] limiting access to those in greatest need. One might assume people would prefer health insurance coverage for a known upcoming medical expense during pregnancy and delivery. But remember that roughly half of pregnancies are unintended and occur among low-income women.[29]

Regardless of pregnancy intention, making enrollment affordable, easier, and consistent may prove beneficial for the childbearing individual and the infant.

Step 3. Coverage: Even if individuals have health insurance, some needed services or providers may not be covered.

Plans have discretion regarding what services and providers they cover. For example, the Affordable Care Act made it possible for individuals to remain on their parents’ health insurance plans until age 26. However, if parents have coverage from a large employer group health plan, the plan may cover the adult child’s prenatal visits but not the adult child’s delivery.[30] Therefore, young adults who stayed on their parents’ plan, in part because of lack of better alternatives or costs, may find themselves with substantial expenses associated with a coverage gap for delivery, yet they may also be ineligible for Medicaid. In another example, short-term health policies, defined as plans with coverage for less than one year, also may not cover maternity care or mental health care.[31]

A related issue centers on the providers covered. Mounting research shows the benefits of birth doulas, yet neither public nor private insurance plans have a requirement to cover doula services.[32] In addition, many mental health providers do not take any form of health insurance, especially Medicaid.[33] Therefore, having enrolled in a health insurance plan may represent a necessary but not sufficient condition to accessing needed providers and services.

Step 4. Choice: Even if services and providers are covered, patients may not be able to choose among plans, institutions, or clinicians, and thus cannot exercise their power in the market to select the care they prefer.

Network adequacy refers to a health plan’s ability to deliver the benefits promised by providing reasonable access to enough in-network providers and all health care services included under the terms of the contract. Yet plans may narrow networks to reduce costs,[34] even if consumers would pay more to have additional choices. Network adequacy may vary based on geographic location.

In addition, roughly three-quarters of employers provide only one option for health insurance coverage, which limits employee choice.[35] The single plan offered may not meet an individual’s needs for mental health or maternity care. Preferences for health plans may vary among individuals with different ethnic and cultural backgrounds. It may also use a high-deductible plan or have limited providers in-network.

Therefore, narrow networks and few, if any, plan choices can lead to another drop in the healthcare quality cascade.

Step 5. Consistency: Even if people have a choice of plan or provider, a consistent source of care may not be accessible.

Although Eisenberg focused on primary care, other forms of care may also prove hard to access consistently. Over one-third of Americans live in a mental health provider shortage area, and approximately 2.2 million women of childbearing age live in maternity care deserts.

The lack of providers and institutions available to provide the care that individuals need causes significant burdens for those who do not have the resources to travel long distances and take time away from other responsibilities such as work or family to make those trips. Workforce shortages in both mental health[36] and obstetrics and gynecology[37] specialties also contribute to this ongoing problem.

Step 6. Referral: Even if care is available and accessible, referral services may not be.

Beyond difficulties regarding network adequacy and shortage areas, lack of referrals may also cause another dip in the quality of perinatal mental health services received. Obstetricians may not feel comfortable making referrals to mental health professionals and may also not feel capable of managing the mental health needs of their patients if a concern arises during a routine exam or screening.[38]

Further challenges can arise postpartum, including a potential gap between referral and follow up. For example, pediatricians can use the Early and Periodic Screening, Diagnostic and Treatment benefit under Medicaid to screen a new mother for depression during her infant’s well-child visit.[39] Yet whether, when, and how pediatricians refer the mother to mental health services if needed remains less than clear.[40]

The transition from obstetric care during pregnancy to primary and mental health care postpartum may leave mothers struggling between visits and providers.

Step 7. Quality: even if people have access to both care and referral services, there may be gaps between the quality of care that can and should be provided and the quality of care that is delivered.

Let’s say a pregnant person has navigated the system and has gained access to insurance, coverage, and providers. Gaps can still exist in the quality of care received.

Knowledge on the quality of perinatal mental health care individuals receive remains opaque given a dearth of information, including relevant quality metrics on prenatal and postpartum care, mental health care, and the intersection of the two. Although the existence of quality metrics does not guarantee acceptable care, when health systems and providers do not face accountability, they may choose to prioritize the care based on mandated ratings.

Several relevant measures assess quality or care provided by public insurance, such as Medicaid,[41] and private insurance, through the Healthcare Effectiveness Data and Information Set (HEDIS).[42] Measures of perinatal care include timeliness of perinatal care, a postpartum visit, low birthweight, and well-child visits within the first 15 months of life. Measures of mental health care include antidepressant medication management, and follow-up after emergency department visits, among a handful of others.[42]

In 2019, the National Center for Quality Assurance introduced two measures of screening for depression during pregnancy and postpartum, respectively. These measures rely on electronic medical records rather than administrative claims data. Measures derived from electronic health records exact a much larger burden for providers, health plans, and health systems. Such measures include assessing whether a provider has screened a patient for a disorder such as depression (regardless of the result of the screening instrument), instead of using billing codes to determine whether a patient had a visit or received a diagnosis. Clinicians rarely use this measure, and they remain underpopulated in health records.

This type of circumstance has played out in the past. In 2004, HEDIS had a measure of optimal provider contacts among individuals who began using an antidepressant medication for depression.[43] When health plans performed poorly on providing optimal visits, defined as three visits, rather than trying to fix the problem, that measure got dropped. That would be like saying to school children that if you do badly enough on a test, we will stop evaluating you on that material. Another measure called the behavioral health risk assessment during pregnancy in Medicaid existed around 2015.[44] Only a handful of states used that measure for a few years, and it also got dropped. These instances created a concern for the future fate of new depression screening measures. In fact, with the new HEDIS measure, less than 20% of pregnant and postpartum individuals received screening.[45]

Even if the two recent perinatal depression screening measures become more widely implemented, screening represents only the beginning of the treatment pathway. Although professional organizations such as American Association of Pediatrics,[46] the American College of Obstetrics and Gynecology,[47] the American Psychiatric Association,[48] and the US Preventive Services Task Force,[49] recommend screening at least once during pregnancy or postpartum, they do not provide consistent guidance on the optimal timing, frequency, and intervals of screening. Screening for depression during the first trimester of pregnancy or the third, two weeks after delivery or two months, may all yield different results with varying recommendations for treatment and follow-up.

Finally, quality measures for perinatal mental health care focus primarily on depression. Yet rates of perinatal anxiety have exceeded those of perinatal depression. Measures exclude other important conditions, including post-traumatic stress disorder, bipolar disorder, schizophrenia, or other traumas. Remaining unaware of other conditions affects the ability to provide robust high-quality perinatal mental health care.

Step 8. Adherence: Even if people receive high-quality care, they may not be adherent to treatment.

The Eisenberg voltage drop model ends with quality of care, but this paper extends the model further. The next step involves patient adherence to treatment. Providers cannot force patients to follow recommended or prescribed treatment regimens, including psychotherapy, psychotropic medication treatment, or a combination of the two. Among people who accept a referral, less than half attend an intake appointment.[50]

Studies show that during pregnancy and postpartum, few people receive evidence-based care, including the Cox et al. cascade model mentioned earlier.[17] They found that less than 10% of individuals with depression during pregnancy or postpartum received adequate treatment and less than 5% achieve remission.

In addition, individuals may discontinue prescribed and evidence-based psychotropic medication regimens before or upon becoming pregnant, either on their own, or because a provider suggests doing so. But data show substantially higher relapse rates among those who discontinue treatment than those who maintain it during pregnancy.[51] The risks and benefits of using psychotropic medications during pregnancy and postpartum versus no medication use remain beyond the scope of this paper. Substantial literature exists on that topic; however, individuals can make their decisions by weighing their own personal preferences and values and in consultation with their provider about health care use during and after pregnancy. One component of the risk-benefit calculation involves weighing the potential outcomes for mother, fetus, and infant with treated and with untreated maternal mental illness.

Although health systems influence quality-of-care patients receive, patients make choices that can affect their likelihood of treatment success.

Step 9. Barriers: Even if people have access to high-quality care and aim to be adherent, societal forces may influence their choices and behaviors.

Next, barriers may influence access to high-quality care that extend beyond treatment adherence. This category considers the impacts of societal forces, such as stigma and racism, on diagnosis, treatment, and outcomes associated with perinatal mental health conditions.

Stigma may influence a person’s willingness to seek or engage in treatment. This can lead to significant gaps between screening and treatment. Stigma can include a perception by others or external stigma and self or internal stigma. Mental health providers also may harbor unconscious bias and stigma toward their own patients, which can negatively affect them.[52] Women want to feel like they are good mothers and blame themselves and feel guilt and shame when their ideals or hopes about pregnancy, delivery, and postpartum do not align with their reality.[53] Such negative thinking can have a harmful effect on maternal-infant bonding and child wellbeing.[54]

Women of color have a higher likelihood of receiving inequitable treatment during pregnancy and postpartum, including heightened surveillance.[55] Structural inequities persist all along the pathway from screening to diagnosis to treatment outcomes.[56] Besides well-documented underdiagnosis and treatment of PMADs in minoritized populations, additional societal barriers decrease the quality-of-care they receive. This includes a lack of education and social support around PMADs, time to get treatment, provider cultural competency, child care, and time for treatment, just to name a few.[57] Earned skepticism and mistrust of providers requires thoughtful attention.[58]

Step 10. Shocks: Even if people had access to high-quality care, limited barriers to adherence or other social obstacles, unanticipated events could disrupt their care pathways.

Finally, external shocks can influence the entire pathway. Even if people have access to high-quality care, limited barriers to adherence or stemming from other social forces, unanticipated events could disrupt perinatal mental health care and associated outcomes. These experiences occur in two major categories, including public health emergencies and natural disasters, such as the COVID-19 pandemic on one hand, and intended and unintended consequences of health policies on the other.

Regarding the pandemic, evidence has rapidly accumulated on the increased prevalence and burden of PMADs over the past several years.[59, 60] Childbearing individuals faced a variety of challenges, in addition to heightened mental health concerns. Lack of access to supportive caregivers or partners during delivery, masking, and isolation all may have had a negative influence on mental health, including a potential inability to afford or stick with treatment.[61] Excessive information seeking and worries about children and medical care also contributed to poor perinatal mental health.[62] Lack of early childcare options also may have contributed to negative delivery and postpartum experiences with mental health conditions.[63] High maternal depression and grief during the pandemic negatively affected maternal-infant bonding.[64]

It will take time to understand the long-term consequences of the pandemic on perinatal mental health. As a society, we ignore the impact of the pandemic at our peril, especially because other rapidly spreading illnesses or viruses will become part of the future.

International Implications

Sources from international literature highlight the importance of PMADs as a leading cause of maternal morbidity and mortality as well.[65, 66] In describing the cascade model, this paper used the US healthcare system as the primary example. However, the model can extend to examine dips in quality-of-care along the perinatal mental health treatment continuum within the international context. Although the US healthcare system may differ from other healthcare systems in many respects, shared commonalities lead to dips in quality-of-care in countries with healthcare systems structured differently than in the US.

One aspect that changes across geographic setting is the cost of healthcare. For example, the general cost of US healthcare spending per capita exceeds spending per capita even compared to other Organization for Economic Cooperation and Developed (OECD) countries.[67] One driver of this cost discrepancy stems from funding of other healthcare systems. To stay with OECD countries as an example, healthcare funding sources in these countries include public taxation, social programs, health insurance, private insurance, and out-of-pocket; on the whole, for-profit insurance companies play a smaller role in other OECD countries than in the US.[68] The US also spends more on healthcare than do other high-income countries (HIC)[69], which in turn spend more on healthcare than low and middle-income countries (LMIC).[70]

However, regardless of variation in structure of healthcare system and the general cost of healthcare, cost-of-illness studies of maternal mental health from other countries indicate substantial costs for this condition no matter where we look. To provide context for the cost estimates that follow, common perinatal mental health conditions have a higher prevalence in LMIC than in HIC.[71] According to a systematic review of literature of UK evidence as well as evidence from other HIC, the lifetime costs of perinatal depression and anxiety were £75,728 and £34,811 per birthing person with depression and anxiety, respectively, across HIC, with a total aggregated cost of £6.6 billion in the UK alone.[72] Across LMIC, the cost of childhood stunting, defined as “permanent harm on children’s development trajectory,” attributed to maternal depression was estimated as USD 14.5 billion in unearned wages.[73] In Brazil, the lifetime cost of perinatal depression and anxiety was estimated as USD 4.86 billion.[74] In South Africa, that cost was estimated as USD 2.8 billion per annual cohort of births, which as a proportion of GDP per capita represents a higher burden than in the UK.[75] Taken together, maternal mental health conditions pose a substantial financial burden worldwide.

As in the US, information about relevant quality metrics within the international context and evidence of enforceable patient safety bundles for mental health globally remains thin to non-existent. One systematic review revealed that the presence of more local evidence surrounding perinatal mental health conditions in HIC than in LMIC; this local evidence, in turn, forms the basis of practice guidelines and policy, so we can expect a commensurate discrepancy in the availability of practice guidelines and policy between HIC and LMIC. [71]

Clinical practice guidelines exist in some international contexts. For example, in the UK, the National Collaborating Centre for Mental Health and National Institute for Healthcare Excellence offer the Perinatal Mental Health Care Pathways guidance[76], and in Australia, the Centre for Perinatal Excellence offers the Mental Health Care in the Perinatal Period: Australian Clinical Practice Guideline.[77] The World Health Organization (WHO) also offers guidance for how best to integrate perinatal mental health services into maternal and child health services.[78] A detailed review of this guidance is beyond the scope of this paper.

One key takeaway from this WHO guidance is the stepped-care approach, which begins with promotion of good perinatal mental health through establishing an environment of respectful and non-stigmatizing care, providing psychoeducation, teaching about stress management, and shoring up social supports, among others, followed by evidence-based treatment of mild perinatal mental health conditions within the context of maternal and child health services and referral of moderate-to-severe mental health conditions to specialists.[78] A recent study reviewed the evidence base for individual and public health-level interventions, including psychotherapy, pharmacotherapy, electroconvulsive therapy (ECT), preconception interventions, and social influencers of mental health such as racism and poverty, offering a nuanced perspective on strength of evidence for each.[79] The authors conclude that reasonable evidence exists for psychological and psychosocial interventions such as psychotherapy, for the use of antidepressants in the postpartum period with appropriate individualized risk-benefit analysis; and that ECT is overall safe for individuals with life-threatening complications such as suicide risk, among others.[79]

Though healthcare systems, cost of illness, and clinical practice differences highlight what may not directly translate to other countries from the quality-of-care cascade model as discussed within the US healthcare context, some aspects of the quality-of-care cascade do cross geographic boundaries and apply to any international context. First, perinatal mental health conditions are common across geographic settings, representing, for example, nearly 20% of medical complications in the perinatal period in LMIC[80], as well as having an estimated prevalence of 6.5% to 12.9% during the perinatal period in HIC.[81] These perinatal mental health conditions affect many people across the globe, and the costs of perinatal mental health remain high worldwide.

Second, although a country’s healthcare system may not rely on for-profit insurers as within the US healthcare system, access to care may remain limited. In LMIC, for example, 76–85% patients who need intervention for maternal mental health conditions do not receive such intervention.[81] Third, a country may lack consistency in availability of health care providers even if a country’s healthcare system does not rely on for-profit insurers as within the US healthcare system. A shortage of appropriate health specialists in LMIC has led to proposals to use large-scale programs utilizing community healthcare workers to fill the gap.[81] On the other hand, in HIC, the integration of perinatal mental health services into maternal and child healthcare services as a way to address consistency, for example, by incorporating screening and some treatments into the work of midwives, has shown economic benefits.[82] Previous studies of mental health interventions such as screening programs, pharmacological treatments, and psychosocial and psychological supports suggest that, overall, these interventions remain cost effective, with cost effectiveness assessed through review of clinical outcomes, quality-adjusted life year gains, or cost-benefit analyses associated with the intervention compared with usual care.[83, 84]

Fourth, barriers to care such as structural inequities and stigma persist globally.[80] Structural inequities and associated differences in social determinants of health present as the discrepancy in prevalence of perinatal mental health conditions between LMIC and HIC described above.[71] Addressing stigma associated with mental health conditions remains essential to the success of addressing maternal depression globally.[85] Related, providers should deliver perinatal mental healthcare in a culturally-sensitive manner appropriate for the country-specific context, yet this may not occur.[82] Finally, shocks such as the global COVID-19 pandemic have exacerbated existing perinatal mental health concerns internationally.[80]

Taken together, regardless of the precise cost of perinatal mental health within a country-specific context, addressing these conditions could have a significant impact on overall cost and quality of healthcare globally. The quality-of-care cascade model presented here could help to identify, understand, and address the complex contributing factors that lead to dips in quality-of-care for perinatal mental health conditions across the world. Policymakers should target these dips in quality-of-care to improve care quality and related economic benefits of appropriate treatment.