Bill Roth, IntegriChain

Welcome to the second installment of the “Popping the Gross-to-Net Bubble” series. In this column, I’ll dive into actionable steps pharmaceutical manufacturers can take to address this issue. This series is not intended as an academic review of the concept or problem but rather as a practical toolset for commercial teams, drawing from my 25 years of experience working on these challenges.

The manufacturer’s role in expanding gross revenue

The concept of gross-to-net (GTN) optimization has been central to the pharmaceutical industry since the mid-to-late 1990s. However, the problem on the net revenue side stems directly from challenges on the gross revenue side, which arguably accelerated in the early 2000s with the rise of specialty pharmaceuticals. While it may be uncomfortable for some to acknowledge, pharma manufacturers have naturally and willingly participated in the inflation of the GTN metric.

First, let’s take a step back. When I launched my consulting practice in late 2001, I noticed several key drivers of pricing decisions.

Manufacturers were experts in their science and in selling to physicians and consumers. However, payers were beginning to flex their influence, and manufacturers were unprepared.

Commercial functions were often siloed, with little collaboration around pricing strategy. Finance and government pricing have typically only been involved to measure the impacts of decisions but is not part of the strategic rationale or decision-making process of how and when discount levels were decided.

Discounting without net revenue strategy.

 Few leaders of manufacturers could effectively articulate the nuances of their discounting strategies. As an example, why was the rebate to a payer 40% and not 37%? The role of market access, which has the biggest impact of the GTN phenomenon, is tasked with formulary positions. Some market access teams had unchecked authority to discount at will. Field sales could establish discounting strategies to provider offices for medical benefit products, as they were the primary influencers of product choice for this type of product. Distributors relied on speculative buying margins, and manufacturers often couldn’t calculate the cost of the distribution channel, but some distributors were able to buy as much as six months of product (pre-Sarbanes Oxley) against double-digit price increases.

At the time, the GTN calculation wasn’t a strategy; it was the measurement of the pricing decisions made by disparate teams, and the role of finance and government pricing was simply to measure the impact of the decisions—not to provide guidance or counsel.

One of our first service offerings was to document all spending and package it in a way that allowed senior leaders, brand managers, and finance teams to understand costs by payer, pharmacy/site of care, distributor, and patient. However, most manufacturers dismissed the need for optimization, relying instead on annual price increases of 9% to 10% in order to maintain margins. It was hard to argue with their logic, as it had worked at the time.

The turning point: Medical benefit and the Medicare Modernization Act

Between 2003 and 2005, everything changed. The precursor documents for the Medicare Modernization Act from the Government Accountability Office and the Congressional Budget Office signaled a shift. The government planned to transition medical benefit products from an average wholesale price-based model to an average selling price (ASP)-based model.

This shift introduced a reimbursement mechanism that required manufacturers to think holistically about their broad discounting strategies and accurately captured discounts. Furthermore, the ASP model, with its two-quarter lag, acted as a natural limiter on price increases. If a manufacturer raised prices beyond 1.5% twice a year, provider offices administering the product could face significant losses, as reimbursement would lag behind the price increase. For products reimbursed at ASP plus 6%, this created a major disincentive for aggressive price hikes.

This change marked the first real example of GTN optimization, and remains the playbook for GTN optimization for medical benefit. Manufacturers began to rationalize discounts to align wholesale acquisition cost (WAC) with ASP. Here’s how it played out.

 For medical benefit products, payers didn’t manage discounts as aggressively as they do now. Discounting to a payer only reduced ASP and, therefore, reimbursement. Removing these discounts was an easy decision. The contemporary view here is very important. While there were 30 or so medical benefit products that received payer discounts about 10 years ago, this number crept into the low 100s in the 2020s with the arrival of biosimilars. Like a siren on the rocks, manufacturers that are lulled into payer discounts on medical benefit products find themselves in a very precarious position with providers. As many biosimilar manufacturers are realizing now, one of the first questions a buy-and-bill provider office will ask is “are you contracted with payers?” If the answer is yes, the opportunity to detail the product is typically over as the provider’s office will understand that the WAC:ASP ratio will likely be off, and the net cost recovery of the product is expected to be negative.

Distributors were already moving away from speculative price increase models. By removing prompt-pay cash discounts and shifting to bona fide fair market value (FMV) service fees, distributors could retain margin without competition—as FMV fees cannot legally be passed on to downstream buyers. This is still an important aspect of managing not only the GTN but the total net value of the sale. Overspending for distribution does not yield any incremental value. The company that spends 10% gets the same, and often less, than the company that spends 1.25%. This can be a little trickier to maneuver, but the use of exclusive distribution networks in specialty or highly limited networks are great tools. With the growth of exclusive distribution networks, with rare exception, all clinics and all hospitals buy from all specialty distributors. This is harder with products with more commoditized products—such as biosimilars and specialty generics—but practical for reference products.

 This was the most challenging area, especially when group purchasing organizations (GPOs) were involved. Manufacturers adopted multi-year strategies, pulling back discounts incrementally by 25-50 basis points per year. While potentially painful in the short term, this proved to be powerful in the long term. This remains the hardest of the areas to optimize GTN, especially if the manufacturer has been discounting aggressively. For buy-and-bill products, however, usually as the WAC:ASP ratio widens, the use of the Assignment of Benefit (AOB) model increases, and the alternative sites of care come into the picture. Unwinding these scenarios are not impossible, but it’s a lot to unwind. Even with a modest differential in the WAC:ASP, it will take roughly 18 months to undue short-term discounting.

As we have now been doing this for products on market and new products, going through this process reveals a critical insight for GTN optimization: cross-functional collaboration was essential. Marketing, field sales, market access, trade, and patient services teams have to work together to develop comprehensive pricing strategies, but unfortunately, these groups don’t naturally align together. Going forward, it is our firm’s strong belief that the best laid out plans need to be completed by new product planning early in the development cycle, either Phase II or very early in Phase III, and finance and government pricing teams need to be in a strategic, not a supporting role, governing these decisions and not simply measuring them. This collaborative approach, while still relatively rare, proves highly successful.

On the pharmacy benefit side, the dynamics are different. Pharmacy benefit managers (PBMs) have consistently consolidated and gained power as drug classes become more and more crowded. Despite price increase caps set by PBMs and government penalties like the Medicaid unit rebate amount consumer price index (CPI-U) penalty (introduced in 1993), manufacturers will often still raise price above the CPI for a given period, take the short-term win, and pay the longer-term penalty.

Not surprisingly, payers, pharmacies, distributors, and GPOs all tolerate and encourage these price increases because they benefit from higher prices. This is exemplified in all of their warnings in the market today related to lowering WACs. It’s obvious to anyone in the industry, but any entity earning a percentage of the list price profits from the inflation. The true burden falls on plan sponsors (governments, employers, and unions) and patients.

The entire industry enables the inflation of GTN. Blaming any one party is nothing more than propaganda and misses the systemic nature of the issue.

In the next installment, I’ll explore the pharmacy benefit landscape in greater detail, examining examples of pricing strategies for first-in-class drugs versus those in mature categories with years of price increases.

For those frustrated by the system, it’s worth noting that several key loopholes have been closed through legislation at both the federal and state levels over the past 15-20 years. I’ll cover these reforms in upcoming columns.

Bill Roth is General Manager and Managing Partner of IntegriChain’s consulting business, which includes Blue Fin Group, a strategy consulting company he started in 2001, and the IntegriChain advisory services business.

Articles

A look at the actionable steps needed to elicit change.


Popping the Gross-to-Net Bubble, Part II: What Manufacturers Can Do About It

This column marks the first of a six-part series on “Popping the Gross-to-Net (GTN) Bubble" in the pharmaceutical industry. The “bubble” refers to the widening gap between the gross revenue (the total sales at list price) and the net revenue (the amount received by the manufacturer after discounts, rebates, fees, and other adjustments). This gap has been expanding significantly in recent years due to changes in regulations surrounding pricing in the industry; expansion of government programs; expansion of the use of programs such as 340B; and increased demands for rebates from pharmacy benefit managers (PBMs) and insurers. It is set to continue to widen further with programs that are still to launch, such as the maximum fair price as part of the Inflation Reduction Act.

My intention of this series isn’t just to point to the bubble, which most pundits see currently as roughly $335 billion to $350 billion and growing. Ultimately, I intend to discuss the future of pharmaceutical pricing and the bubble, innovative ideas to address the bubble, how some therapeutic areas and certain product archetypes and classes are more or less sensitive to the bubble, who wins and loses if the bubble grows or pops/deflates, and mechanically, how rebates and discounts fuel the bubble. For this first column, we should simply begin by understanding the bubble itself.

When someone in our industry refers to the gross sales, this refers to the wholesale acquisition cost (WAC) or list price of a product times the units sold in a given period of time. As the vast majority of rebates, discounts, fees, or other price concessions begin with the WAC, this is our starting point of the calculation.

From there, we start to see rebates, discounts, and other price concessions offered. I always like looking at the price concessions by payer type and by physical channel. It gives a weighted view to the flow of the dollar. This view also allows baseline visibility to where there can be duplicate discounts and other areas of potential overpayments. The trick to understanding the baseline numbers are to tease apart which are statutory and required, which are negotiable, and which ones are truly discretionary. The default view by many in the industry is that these rebates and discounts are intended to secure a more favorable position on insurance formularies, making the drug more accessible to patients but reducing the actual revenue manufacturers receive. The reality can be somewhat different.

This is the amount that pharmaceutical manufacturers actually retain after these rebates, discounts, and other adjustments. When publicly traded manufacturers disclose their revenues, they always speak in terms of net revenues, but rarely do they disclose the full walk-down of all the price concessions that lead to the net revenue number.

The expanding gross-to-net bubble has significant implications for the industry. While it sounds like a good idea to “pop” the bubble, most payers, providers, pharmacies, group purchasing organizations, and distributors and their business models depend on the high WAC and the low cost. So, knowing the impacts of the deployment of any strategy are key to any successful initiative around a gross-to-net improvement. What’s maddening at the moment is the fact that more drugs that are coming to market begin with a significant reduction from GTN and attempt to make improvements over time. The challenge is that many of the pricing and price change options have been recently constrained. Long-range financial planning is now crucial to the success of managing this.

The term bubble implies that the situation is unsustainable and may eventually burst due to rising public and regulatory scrutiny. Policymakers are increasingly concerned that the system is inefficient and opaque, leading to discussions around price reforms and more transparent pricing models.

Neck-in-neck with the challenges of gaining product approval, the gross-to-net bubble is the key financial hurdle and hardship facing the pharma industry. Products and life sciences organizations that cannot adequately contend
with this issue face product or even company discontinuation.

Certain types of pharmaceuticals are indeed more impacted by the gross-to-net bubble than others. Generally, high-cost brand-name drugs, especially those in competitive therapeutic areas, face the largest GTN disparities. Here’s a breakdown of the categories most affected.

Drugs for chronic conditions such as diabetes (e.g., insulin), cardiovascular diseases, and asthma are particularly affected because they are commonly prescribed and often require ongoing, long-term use. Manufacturers of these drugs offer significant rebates to ensure their products are included on formularies, as payers seek to manage the high costs associated with chronic care.

 These high-cost medications for complex conditions—such as rheumatoid arthritis, oncology, and multiple sclerosis—tend to see large GTN gaps. Due to intense competition and high list prices, these drugs typically have substantial rebates, discounts, and other concessions, especially given that many are administered under healthcare professional oversight and subject to payer negotiations.

 For drugs in areas with multiple treatment options—such as rheumatoid arthritis or cholesterol management—manufacturers often offer large rebates to secure preferred positions on PBM and insurer formularies. The competition drives aggressive discounting, leading to significant erosion between gross and net revenue.

 Drugs heavily impacted by the Medicaid rebate program or the 340B Drug Pricing Program experience large reductions in net revenue. Under Medicaid, mandatory rebates often exceed 23% of the drug's average manufacturer price, with additional discounts if the drug’s price rises faster than inflation. The 340B Drug Pricing Program also mandates that certain healthcare providers receive discounted drug prices, significantly reducing revenue for manufacturers in these channels.

 Although not traditionally considered under the GTN bubble, biosimilars are increasingly influenced by these pressures. To compete with originator biologics, biosimilar manufacturers offer significant discounts, which further compounds the bubble effect in the biologic space.

After spending more than 30 years in the industry and working with a few hundred manufacturers across the majority of therapeutic areas and drug classes, I see a few main reasons why the pharmaceuticals manufacturer struggle with the concept and, consequently, the strategies deployed. The first issue is the siloes in the commercialization group. There are five main commercial functions and two financial functions that commonly don’t align well. If the marketing, field sales, payer sales, channel sales, patient access and affordability, finance, and government pricing teams are part of a company with more than just a few products, the chances of either these seven teams understanding one another or reporting into a combined function with full P&L responsibility are all very low.

Stay tuned in future issues for the next column in this series where I’ll explore the types of rebates and discounts, and how these fuel the bubble. This will also lay the foundation for where the bubble can be mitigated and how to improve net revenues. 

A crash course on the bubble itself, including therapeutic areas most affected.

Going Inside the Gross-to-Net Bubble and Its Nuances

, there was one main concern on our minds: how to launch a general medicine brand into a progressively changing and fragmenting retail and ecommerce landscape. The historical model of “stick it in wholesale and pull through retail” clearly no longer worked. Almost all the brands launched into this model failed to meet their forecasts. At Blue Fin Group, we received call after call from our manufacturer clients complaining that branded prescriptions were increasingly being turned away by retail pharmacies of all kinds, with estimated abandonment rates reaching 50% to 60%. These were products with strong formulary coverage and reasonable affordability profiles for all forms of payer sponsored lives.

We believed the first signs of this issue became evident with the struggles faced by medications in the specialty lite category: product priced above $650 or below $2,000 a month. A solid example was the original pricing of PCSK9 inhibitors launched in 2015. Our attention at the time was solidly on this specialty lite area, but that product archetype simply proved to be the canary in the coal mine. The reasons for the struggle were that similar access hurdles such as prior authorizations and step edits were driving up the cost to serve, all while reimbursement on the products was in decline and cost of goods was rising.

With the arrival of the general medicine patent cliff came in late 2016, the profitable generic wave machine ended. And adding insult to injury, the government's CPI-U penalty to the generics started on Jan. 1, 2017, effectively ending the ability for generics to increase price after their initial race to the bottom for the first 18 months post-generic launch. We were concerned that this problem could spiral out of control, and we were the only firm talking about this online and on the industry conference stage. In fact, I presented this problem at the Informa Connect Trade & Channel Strategies meeting, which I’ve spoken at for the last 18 years. People in the audience had a look on their faces as if they had no idea what I was speaking about. Unfortunately, our fears and concerns were realized.

Over the past eight years, the situation has not only intensified but has also expanded to include regular branded drugs that were once readily available at every retail pharmacy across the country. Currently, there are about 4,400 branded prescription drugs that comprise a mere 1.5% of all prescriptions dispensed by retail pharmacies. The other 100 brands account for another 6.5% of all prescriptions, with the remainder being generics.

We’ve tried shaking and waking up industry to alert them to the problem. I’d have conversations with retail pharmacy leaders asking us to help educate pharma about the problem. None of the other industry pundits picked up on the gravitas of the situation. And when the direct and indirect renumeration fees went from the back end of the prescription adjudication and reimbursement process, we knew it was going to be like throwing a bucket of gas on an already burning fire.

I credit writing this series for making me realize that, in fact, this started long before specialty lite and the patent cliff. A key event that marked a shift in the classic insurance-based retail pharmacy market (see Figure 1 below) was the reintroduction of cash pharmacy back in 2006 with Walmart’s $4 generics. Most other major retailers followed suit. When 2011 rolled around and we witnessed the launch of GoodRx, I honestly didn’t think much of it. My eye was off the generic ball. I was centered on brands and mainly specialty brands at that. What GoodRx did in hindsight was democratize the prescription. It was the first tool a consumer could use to shop pharmacies and prices. For generics, this was a big deal. Two years after the patent cliff in 2018, the e-commerce pharmacy landscape changed dramatically with the arrival of Amazon Pharmacy, followed by its acquisition of PillPack. Arguably the pandemic masked this issue for a couple years from 2020 to 2022 until the shark-tank minded Mark Cuban’s Cost Plus Drugs further disrupted the market in 2022. 

From 2015 through current day, we have seen the increasing prevalence of digital pharmacies, which many manufacturers hoped would function like specialty pharmacies for retail. That model, however, hasn’t proven to deliver any better net revenue effect than retail from the data points we’ve studied with clients. It’s an alternative, but not necessarily a better one. Then, in the 2020 timeframe, came the emergence of tech-enabled hubs, a retail affordable solution that works somewhat like the patient services platforms used for specialty.

So the crux of my series on “Who Moved My Prescription?” parities the dilemma in the 1990s book 

 in which every day the mouse comes out of the hole and gets the same cheese, and without explanation, one day, the cheese is simply not there. If everything was working fine, the likes of Eli Lilly and Pfizer wouldn’t be launching direct-to-patient models. And now we are contending with telehealth/ecommerce/cash models that we refer to as lifestyle pharmacies, such as Ro and Hims/Hers. The illustration alluded to above shows the chaos and complexity of the current state of retail.

Today, we’re seeing vastly different approaches and perspectives on the profession of pharmacy, and it’s clear that neither the industry nor drug manufacturers know exactly where to turn. The old guard is trying to reinvent themselves a little too late, and the new guard is hoping to come in and blow it all up. Many manufacturer leaders are hoping this issue will resolve itself or that a new model will emerge to address these challenges.

From working with more than 450 manufacturers, we get a unique view of the industry and we see that branded Rx makers are left with a puzzle. It’s not about signaling the downfall of retail, the rise of e-commerce, or the increasing prominence of cash. Nor is it about promoting new pharmacy models.

Over the three decades I’ve been in this industry, we have taken pharmacies for granted. I’ve actually heard retail pharmacy called a “dumb pipe.” We focused on the science, the doctors, the payers, and the patients. We cared about specialty pharmacy because we could see the prescription journey and tie that back to our own patient services platforms. Specialty was cool. Retail was and is a black box.

So where do we go from here? We educate ourselves. We listen. We seek to understand. We try new things. We experiment. We educate management teams, boards, and investors. We reinvent.

Thank you for sticking with me throughout this series. I’ve learned and reflected a lot in the process. I’m excited to share there I’ve been invited back for another series in 2025; there are still a few more topics that need our attention.

Why this trend is occurring and ways that drugmakers are pivoting in order to remedy the situation.

Who Moved My Prescription? Part VI: Abandonment of Branded Prescriptions

On June 27, 2024, Walgreens announced that it was reviewing 2,100 of its 8,600 pharmacies in the US for performance and likely closing of 700 to 800 pharmacies. CVS—which reported that it had 9,939 stores in 2021—has closed almost 700 since then. The company now has 9,273 stores, and plans to close another 300 in 2024. Rite Aid, in 2021, had 2,447 stores and now has 1,704, according to the organization's latest reports.

While independents are holding on at 19,400 stores, according to the National Community Pharmacy Association, as many as a third of them are either planning to close their pharmacies or converting to non-insurance cash-pay pharmacies and focusing on generics.

Several factors have contributed to the wave of pharmacy store closings across the US.

. Reduced reimbursement rates from insurance companies—mixed with rising labor costs, store theft, rising cost of goods, and the high cost of maintaining a brick-and-mortar presence—have made it difficult for many pharmacies to stay afloat.

What’s bad for one may be good for another. Companies such as Amazon Pharmacy; Mark Cuban Cost Plus Drugs; Alto Pharmacy; Capsule; lifestyle pharmacies such as Hims, Hers, Roman, and Lemonaid; cash pharmacies including Blueberry and Eagle; and multi-model pharmacies such as OneroRx and others have come forward to disrupt the world of high-volume, low-cost generic brick-and-mortar pharmacy.

The rise of these new forms of pharmacies and prescription services have changed consumer behavior and reduced foot traffic to physical pharmacy stores. This shift has been accelerated by the COVID-19 pandemic, as more people turned to online options for their medication needs.

Retail pharmacies have become very lazy over the last 20 years as they have outsourced their interactions with payers to pharmacy services administrative organizations and outsourced their sourcing of product to wholesalers. They are told not to pay attention to the losses on individual products and instead, focus on the all-in number. But that’s not working anymore. Otherwise, we wouldn’t be seeing these store closures.

In recent years, the phenomenon of “pharmacy deserts” has gained increasing attention. A pharmacy desert is an area where residents have limited access to pharmaceutical services and medications due to the absence of nearby pharmacies. This issue has been exacerbated by the closure of numerous pharmacy stores, particularly in rural and low-income urban areas.

The closure of pharmacy stores and the resulting pharmacy deserts have profound implications for public health and community well-being.

Residents in pharmacy deserts often struggle to obtain necessary medications. This lack of access can lead to poor health outcomes, particularly for individuals with chronic conditions such as diabetes, hypertension, and asthma.

 Pharmacy deserts disproportionately affect low-income and minority communities, exacerbating existing health disparities. These populations are already at higher risk for certain health conditions and may have fewer resources to seek alternative sources of medication.

 Limited access to pharmacies can lead to increased healthcare costs. When people cannot access their medications, they are more likely to experience complications that require emergency department visits or hospitalizations, placing additional strain on the healthcare system.

 Pharmacies often provide essential health services beyond medication dispensing, such as immunizations, health screenings, and consultations. The loss of these services can leave significant gaps in community healthcare, particularly in areas with limited access to other healthcare providers.

While the problem of pharmacy deserts is already receiving attention in several states, it doesn’t appear that states can prop up a business model in decline, no matter how savvy it is or how involved policymakers, healthcare providers, and community organizations may get.

. Community organizations and local governments can play a crucial role in identifying and addressing pharmacy deserts. Initiatives such as mobile pharmacies, community health clinics, and partnerships with local businesses can help provide alternative sources of medication and healthcare services.

. It may not be enough to shore up new forms of pharmacy for these geographies. Therefore, battle lines are drawing between brick-and-mortar and mail order/home delivery models. And recall that rural and low-income patients who are likely to be disproportionately affected may not practically be able to access these.

. Sorry, this isn’t real. It’s just the No. 1 response we hear from pharmaceutical manufacturers. It seems logical, but recall that these are the entities that have swelled national drug code blocks from 30 to more than 750 in the last 10 years, the ones that added direct and indirect remunerations to non-Medicare volume and even lowered rates for 2025. With that said, pharmacies believe they won something when National Average Drug Acquisition Cost (NADAC) was implemented for Medicaid, but it hasn’t been the silver bullet they were hoping for. It stemmed some losses but didn’t save pharmacy. From research that Blue Fin Group has conducted in the market in 2023 and 2024, payers are promoting to retailers that unprofitable prescriptions can be sent to the payers’ mail order businesses. So why would the payers solve it? Why not let retail implode and vertically stack the business back to mail order, which is exactly what the payers have been doing in specialty pharmacy with their business, especially in the areas of immunology, neurology, dermatology, and respiratory.

If you look up polypharmacy, the actual definition is a patient who’s on five or more medications. The way I’ve used the word throughout my career is a patient who has access to multiple pharmacies for their medications. It wouldn’t be hard to argue that our industry did this to itself with the rise of specialty pharmacy over the last 20 or so years. In the late 1990s, specialty products didn’t fit retail pharmacy. The products were too expensive, unit demand too little, and they came with costly education, prior authorization support, step edits, reauthorization, and very challenging patient economics. It took a new model of pharmacy to support this. I wonder if that’s what we’re seeing here now in the retail space.

The retail pharmacy market is essentially generics and top-100 brands, but everything else is splintering into direct-to-patient models or one-offs thinking that non-specialty products can be forced to work like specialty.

In a custom analysis, our firm Blue Fin Group has done of all the drug launches since the beginning of 2023, and the verdict at the moment is that despite all the attempts at digital pharmacy or any model that is an end-around to the pharmacy benefits manager model, none prove to be successful. But it is clear that we’re going the wrong direction. We are splintering care. We are creating an environment of mixed insurance and cash business. We are risking contra-indications and drug interactions, let alone adherence to therapies.

In summary, I wish I had better news to share. The reality is that retail pharmacy doesn’t know a way out of this situation. My personal assumption is that pharmacies will seek to improve their cost of goods sold by going direct when and where it makes sense. My firm sees the pharmacies trading data and inventory guarantees and becoming part of limited brick-and-mortar and home delivery networks that operate in concert with tech-enabled hubs. The hope is that these pharmacies would then return to profitability, solve for pharmacy deserts with their own mail order or home delivery models, and strike a balance of working with cash pay pharmacy models to find that new model.

Stay tuned for the next and final column in this six-part series where I wrap it all up for 2024 and set our sights on the hottest topic of 2025.

Bill Roth is the SVP of IntegriChain’s consulting business, which includes Blue Fin Group, a strategy consulting company he started in 2001, and the IntegriChain advisory services business.

Making sense of the accelerating trends and challenges—including store closings, pharmacy deserts, and the rise of polypharmacy.