Regional Disparities in Qualified Health Plans’ Prior Authorization Requirements for HIV Pre-exposure Prophylaxis in the United States

This cross-sectional study investigates whether there are regional disparities in prior authorization requirements for combined tenofovir disoproxil fumarate and emtricitabine with qualified health plans and assesses if any qualified health plan characteristics explain the disparities.


Introduction
Biomedical HIV prevention with HIV pre-exposure prophylaxis (PrEP) has been available since 2012 and, if widely used, may help end the HIV epidemic in the United States. [1][2][3] The efficacy of PrEP in reducing HIV acquisition exceeds 90% for sexual encounters and exceeds 70% for injection drug use. 4 Averting 1 new HIV transmission saves the health care system more than $400 000 in lifetime costs. 5 However, PrEP use lags behind the HIV epidemic, particularly in the South. [6][7][8] The most frequently cited barriers in the South for low PrEP use include lack of clinicians with PrEP knowledge, absence of health insurance, stigma, and underestimation of personal HIV risk. [9][10][11][12][13] Improved HIV biomedical prevention through access to PrEP is 1 of the essential components of the federal government's Ending the HIV Epidemic initiative. 3 Although high costs [14][15][16] and lack of health insurance 17 are barriers to PrEP use, we also considered whether there may be regional differences in qualified health plans' (QHPs') benefit design associated with disparities in access to PrEP. QHPs are health insurance plans that are certified by the Affordable Care Act (ACA) health insurance marketplaces and meet federal criteria. 18 With more than 11 million Americans relying on QHPs for health care access, any disparities in QHPs' benefit design for PrEP access could have far-reaching effects. 19 In fact, the Centers for Disease Control and Prevention (CDC) estimates that two-thirds of men who have sex with men, for whom PrEP is indicated, have private insurance. 20 Instances of potential QHP discriminatory benefit design for HIV treatment (eg, placing all or most antiretroviral medications for HIV treatment on the highest specialty tier) 21-23 have prompted concern that there may be similar plan design challenges affecting access to PrEP. Given that primary care clinicians and HIV clinicians cite prior authorization requirement as 1 of the main barriers to prescribing PrEP, 24,25 we focused on this aspect of formulary design.
As part of utilization management, insurance companies can require prior authorization, which involves the clinician or office staff obtaining prior approval from the insurance company for the coverage of prescription medication. Through this process, clinicians must justify the medication as medically necessary and may be asked to document that the patient meets specified clinical criteria.
This process often needs to occur before a patient can receive the medication from the pharmacy. 26 Prior authorization requirements have been shown to reduce both necessary and unnecessary medication use. 27,28 The objective of this study was to assess whether there are regional differences in QHPs' use of prior authorization requirements for combined tenofovir disoproxil fumarate and emtricitabine that may contribute to the decreased PrEP uptake in the South. In addition, we sought to assess whether any QHP characteristics could explain regional disparities in prior authorization requirements. Virginia). 30 The following plan and rating area variables were included in analyses: national issuer, high deductible, PrEP cost-sharing structure, PrEP specialty drug tier status, plan level, rating area urbanicity, and rating area competition. The QHP issuer was categorized as national or regional. The issuer was defined as national if the company offered at least 1 plan in all 4 regions of the United States. A QHP was categorized as having a high deductible or low deductible using the Internal Revenue Service's 2019 definition of a high deductible 31 as greater than $1350. Using the formularies, each QHP's PrEP cost-sharing structure was categorized as coinsurance or copay, and the QHP's handling of PrEP was assessed as specialty drug tier or nonspecialty drug tier. Plan level, also called metal level, reflects the actuarial value (AV) of a plan, or how persons and their insurance company share costs. The higher the AV, the more generous the plan is. Catastrophic coverage has the lowest AV, followed by bronze, silver, gold, and platinum, which has the highest AV. 32 Therefore, plan level was categorized as platinum, gold, silver, bronze, or catastrophic. Plan rating areas were defined as urban if the mean urbanicity score of the rating areas' counties was less than 5 as determined by the National Center for Health Statistics. 33 Rating area competition was quantified as the number of issuers offering QHPs in a rating area.

Statistical Analysis
Descriptive statistics were used for prior authorization requirement and plan characteristics.
Categorical plan characteristics were compared between regions using the χ 2 test, and the Kruskal-Wallis test was used for comparing the mean number of issuers between regions.
Log-binomial regression was used to estimate the primary outcomes of the association between region and prior authorization requirement and the associations between plan characteristics and prior authorization requirement, adjusting for region. The adjusted model was used to assess whether other plan characteristics (national issuer, high deductible, PrEP cost-sharing structure, PrEP specialty drug tier status, plan level, rating area urbanicity, and rating area competition) may explain the disparities in prior authorization requirement by region. Because the association between high deductible and prior authorization requirement differed by region, an interaction term for region and high deductible was included.
A secondary outcome was mapping the percentage of QHPs with prior authorization requirements for PrEP by rating area. Analyses were performed using R (R Foundation for Statistical Computing) and RStudio (RStudio, Inc). P values were 2-sided, with P < .05 used as the threshold for statistical significance.
Plan characteristics differed by region ( Table 1). The following variables are highlighted because they share a similar regional pattern to the regional pattern of prior authorization requirement. Plans in the South were most likely to be offered by a national issuer (17.0%), followed by the Midwest (14.0%), the West (9.6%), and the Northeast (8.9%). Furthermore, plans in the South were offered in rating areas with a mean of 4.3 fewer issuers than the Northeast, whereas the Midwest and West had 2.6 and 2.1 fewer distinct issuers than the Northeast, respectively.  Figure 2 shows the percentage of QHPs that required PrEP prior authorization in each rating area.

Multivariable Model for Prior Authorization Requirement
Although there were regional disparities in plan characteristics and some plan characteristics were associated with prior authorization requirement, region was still strongly associated with prior authorization requirement when accounting for plan characteristics ( In the adjusted model, plans offered by national companies were 3.33 (95% CI, 3.07-3.61) times as likely as plans offered by regional companies to require prior authorization (

Discussion
The QHPs in the southern United States were more likely than QHPs in other regions of the United States to require prior authorization for combined tenofovir disoproxil fumarate and emtricitabine. In addition, this disparity could not be explained by differences in other plan characteristics. This finding is concerning for possible discriminatory benefit design (benefit design that prevents or delays people with complex or expensive conditions from obtaining appropriate treatment) because prior authorization is being used differently depending on the QHP's region.
Arguably, the South is the region that is most in need of access to PrEP given that the South has the highest number of annual new HIV diagnoses and has lower PrEP use than the rest of the country. [6][7][8]34 More than half of the African American population in the United States live in the South. 35 The CDC has estimated that the lifetime risk of acquiring HIV in the United States is 1:22 for African American men, 1:54 for African American women, and an astounding 1:2 for African American men who have sex with men. 36 Moreover, the South has higher rates of stigma and bias associated with HIV and the lesbian, gay, bisexual, transgender, and queer or questioning communities as well as higher rates of HIV criminalization laws, which create additional barriers to PrEP uptake in the South. 13,[37][38][39][40]  Prior authorizations are often used when there is more than 1 interchangeable therapy available within a drug class and the cost differs. 41 Until recently, there was only 1 medication approved by the US Food and Drug Administration for PrEP, meaning that prior authorization was used for other reasons, for instance, to determine HIV risk or clinical eligibility for PrEP. There have been instances of insurance companies denying PrEP prior authorization requests for discriminatory reasons. 42 In addition, the prior authorization process has been found to be frustrating to patients and clinicians because of the complexity, the consequences of delayed or denied medications, and poor communication between stakeholders. 43,44 In a recent survey, 92% of physicians reported that prior authorizations lead to delays in medically necessary therapy, and 78% of physicians reported that a prior authorization at least sometimes leads to patients' abandoning a recommended medication. 45  coverage requirement of PrEP may lead to insurers trying to limit access by using arbitrary, nonclinically based prior authorization requirements. 49 Unfortunately, our research demonstrates that prior authorization is already being used by almost 20% of QHPs and that it is being used more often in the South.
Plan characteristics did not explain the differences in prior authorization use among the regions.
The QHPs in the South are more likely to be offered by a national company, and approximately 50% of QHPs offered by national issuers require prior authorization for PrEP. This finding may explain part of the higher rate in the South, but it does not completely explain the disparity. The ACA includes nondiscrimination requirements that prohibit issuers from using arbitrary plan designs limiting access to care, such as adverse tiering that places all or most medications used to treat a specific condition on a specialty tier or excessive prior authorization. 18 Regulators review issuers that are outliers in their use of prior authorization. 50 However, this type of outlier analysis would not identify Decreased competition in the marketplace (eg, issuers discontinuing plans sold in the marketplace) has been associated with changes in QHP premiums. [51][52][53] Given this association, we anticipated that lack of competition could be playing a role in issuers' decisions about the PrEP prior authorization requirement, with areas with lower competition using prior authorization more.
Although rating area competition as measured by the number of issuers per rating area was lowest in the South, it was not found to be associated with QHPs' prior authorization requirement in the adjusted model.
Some QHP benefit design factors that shift drug costs to the consumer, such as coinsurance cost-sharing, specialty drug tiering, and catastrophic-level plans, were associated with lower rates of prior authorization requirement for combined tenofovir disoproxil fumarate and emtricitabine.
High-deductible plans also shift cost to consumers and were associated with lower rates of prior authorization in all regions except for the South. Although many of the factors that shift cost to the consumer were associated with less use of prior authorization, these factors may also adversely affect PrEP access because studies 54,55 have demonstrated that shifting pharmacy costs to consumers results in less medication use. In addition, 8% of the US adult population report prescription medication nonadherence because of prescription drug costs. 56 Decreased PrEP adherence can result in nontherapeutic levels and higher risk for acquiring HIV, especially for people having vaginal intercourse. 57,58 Limitations This study has limitations. One limitation is that we could not assess the association of prior authorization requirements with access to PrEP. Also, because we did not have access to PrEP need or prescriptions at the rating area level, we could not assess whether PrEP need or PrEP prescriptions are associated with the rate of QHPs requiring prior authorization in a given rating area. Future work should explore the association of prior authorization with prescribing PrEP, filling PrEP prescriptions, and PrEP persistence. Future research will need to examine if and how the availability of multiple formulations of PrEP would affect access, the association of state-level policies with plan prior authorization use, and whether the use of pharmacy benefit managers influences PrEP prior authorization requirements.