The Lilly Digest

Today we discuss the history of the storied institution now known as the NCPA Digest.

“Every survey is an open target for critics and the studies discussed in this article are no exception. For example, there are those who feel that the Lilly Digest cost and profit averages represent those of "above average" pharmacies. Even if this contention should be proved, and up to this point all such statements have been speculative, it can be said without fear of contradiction that this survey, and others as well, fulfills a most useful purpose in providing pharmacy proprietors and managers with goals to achieve and guideposts to follow.” - commentary on the 1954 Lilly Digest.

I have read Adam Fein’s Drug Channels blog for about a decade now. While I often disagree with his conclusions, I do appreciate that he bases his analysis on factual sources. I think that by reading the (helpfully linked!) source material in his blog, a person can become quite well-informed about the pharmacy industry.

His post this Tuesday was his annual breakdown of the NCPA Digest. I read this post each year with interest, because I’ve submitted the family pharmacy’s data to the Digest almost every year for the past 10 years. I’d like to add a few comments about the NCPA Digest, as well as some points that I made on LinkedIn regarding his post.

The NCPA Digest is a long running publication that dates back to 1932. It was published as the Lilly Digest by Eli Lilly and Company from 1932 to 1992. NCPA (then called NARD – the National Association of Retail Druggists) took over the Digest in 1995 and has partnered with many different industry partners (Searle, Pharmacia, Pfizer, Cardinal) to fund the operation of the Digest. It therefore provides some FASCINATING comparisons over such a long time series.

For example, “In the compounding period of 1941, the average pharmacy filled about 5,000 prescriptions a year, approximately 15 a day.” I can’t imagine a typical pharmacy filling 15 prescriptions total per day. But I bet my grandpa could. I was recently given an old prescription bottle from January 6, 1958 dispensed by King-Jolley Rexall Drug, my grandfather’s pharmacy. It bears a prescription number of 31766. My grandpa opened his doors on June 7, 1954, so over his first 3.5 years in business, he dispensed an average of 8857 prescriptions/year. The 1957 Lilly Digest reveals that that’s below average - the typical survey-respondent pharmacy did 13502 prescriptions/year.

Grandpa’s 31,766th prescription. Inside are red tablets with no markings.

Anyway – on to the current issue of the Digest (which is the 2021 Digest that summarizes data from 2020 operations). I think that understanding what the Digest is and isn’t is helpful here. The Digest is a summary of voluntarily submitted data from pharmacy owners based on their internal financial statements and management statistics. It is NOT an audited 10k annual report filing of the results of operations of the mega-corporation Family Pharmacies, Inc crowing its growth in its GAAP EBITDA. If you’ve ever tried to buy a small business, you can understand the difference. Pharmacy owners have different bookkeeping systems (for the record, wholesaler rebates are properly an adjustment to COGS, they’re not revenue) and often include personal expenses in their financials (is your cell phone a company perk or a personal expense?)

The survey attempts to account for the differences in financial reports between pharmacies by asking specific questions about the major points of contention in accounting practices (how do you categorize your rebates? How much were they?). It’s a well-thought through survey document, but I still have uncertainties about how to answer some of the questions – a part asks how many prescriptions were filled with brands and how many with generics. Given that 1/3 of our volume is custom compounded prescriptions, there are a few ways I can answer that – compounds can arguably be included with brands, included with generics, or excluded from the answer entirely. Each choice has bearing on the final result that Drug Channels calls out - “IQVIA data show that for 2020, the GDR for unbranded generics in the overall market was 88.5%. The NCPA Digest reports that the GDR for independent pharmacies was only 86% for 2020.” Including compounds with brands will understate the GDR, with generics will overstate the GDR, and excluding compounds will deflate the volume of the pharmacy, increasing the calculated cost of dispensing.

I’ll note here that I wouldn’t be surprised if Drug Channels is identifying a true phenomenon about relative generic dispensing volumes, but if current trends continue, I think that independent pharmacies will very rapidly approach 100% GDR (current estimate – by 2024 most indies won’t stock ANY branded medications [80% confidence], and you’ll see the same thing at chains – [confidence 40%]). A few factors lead me to these conclusions. 1) Most PBMs (not all!) offer substantially lower reimbursement for 90 day supplies of brands than 30 day supplies of brands. Say for example – AWP minus X% for 30 days vs. AWP minus Y%, where Y = X+3. A Homo economicus pharmacy operator with perfect knowledge (hard to come by with DIR fees and wholesaler rebates obfuscating the true economics of a transaction!) will tend to dispense branded drugs for 30 day supplies preferentially due to this fact alone. It’s also an active point of contention among pharmacists as to whether 90 (or 100) day supplies are objectively beneficial compared to 30 day supplies (see here, for example). Taken together, this will tend to depress raw prescription count GDRs. Notably, I don’t believe that the Digest attempts to account for 30-day equivalent prescriptions like Walgreens and CVS do in their 10-k filings. The IQVIA survey DOES appear to adjust to 30-day equivalents.

In summary, I believe that the difference in GDR noted by Dr. Fein is due to 1) 30-day equivalent adjustment 2) some element of inconsistent reporting by owners and 3) laughably terrible 90 day reimbursement terms (outside of 340b, there’s no such thing as buying brands at 10% below NADAC… but PBMs subscribe to Rule #2, so I guess economic realities need not apply).

A couple of final things I’ll note for my pharmacy owner readers – 1) I think it’s useful to compare your own financials to the figures in the Digest and to the figures in the major chain pharmacies’ 10-k filings, divided by their number of pharmacies/prescriptions. 2) I think that Dr. Fein’s comparison of Owner’s Discretionary Profit to average pharmacist salary is useful.

Regarding point 1, “In 2020, the average per-prescription revenue in the NCPA sample was $55.96.” Compare CVS/Health's 2020 10-k filing stating that they filled 1456.2M prescriptions for $70176M, or 70176/1456.2=$47.89/prescription in their Retail/LTC division (doesn't include specialty, so the smaller per rx figure makes sense to me). I presume that their fill-at-specialty-pickup-at-retail program counts these rx revenues in the specialty pharmacy division.

Compare Walgreens as well, page 45 of their 10-k shows $107701M/818M rxs = $131.66/rx - they don't appear to report their specialty pharmacies as a separate division. I’ll also note that I’m not certain that the “Pharmacy” revenue is ONLY prescription fills. In other words, the $55.96 Digest per prescription revenue figure is believable (it’s also down about $6 since I started paying attention to the Digest, hence my annoyance at people that blame pharmacists for “skyrocketing drug prices.”)

Regarding point 2, if a pharmacist operates their own pharmacy, and pays themselves the average salary of $124,000, then the pharmacy’s net profit is ~$34,000/year, about 1% of typical sales or <$0.50/prescription transaction. (This is REALLY depressing when compared to the 2009 digest, which showed an owner’s discretionary profit of ~$274k - a whole salary bigger than the 2020 figure of ~$158k, and that doesn’t account for 11 years of inflation).

Also, by focusing on a single year’s digest, Dr. Fein seems to miss the big picture. In both his 2011 article and his article from yesterday, he makes the same point “pharmacy gross profits remain stable,” - but the figure in 2009 was 23.8% vs. 21.9% in 2020. A 1.8% decrease in gross profit margin may not sound like a lot, but it adds up to quite a lot especially when you add in the $6 lower sales price per prescription, and the decrease in volume from 64635 prescriptions/year to 57648/year, it adds up to ~$250,000/year in decreased gross margin for the average pharmacy in nominal terms - accounting for inflation it’s about $300,000/year of margin that the average pharmacy has lost. That’s not “stable gross margins.” That’s where all the “cushion money” has gone, folks.

Dr. Fein concludes his analysis with this statement, “For more than 10 years, I’ve been publishing reviews of independent pharmacies’ economics. My message to pharmacy owners has been simple: Get big, get focused, or get out. If a small pharmacy wants to win in today’s consolidating drug channel, it needs scale or differentiation.” I’d like to opine on this statement for a moment – First, to quote Luke Slindee, “These days, simply answering the phone provides differentiation from the big chains.” Pharmacy appears to have reached peak commoditization. It’s hard to imagine a more commoditized service than one where just answering the phone differentiates you, but that’s where we’re at. Second, I think that the point is well-taken – if you want to succeed in pharmacy, you need to do things differently than they’ve been done in the past. To me this means doubling down on the professional education we received and figuring out how to charge primarily for the knowledge in our heads and secondarily for the pills on our shelves. Answering the phone is a good start.

Regarding “get big” – from what I observe, the benefits of scale under common ownership peter out REALLY quickly. You can reach pretty much optimal cost of goods with the right mix of products, a buying group, and about $10-20M in purchases per year – that’s like 5 pharmacies. Past that point, scale tends to increase losses to bureaucracy faster than it decreases cost of goods. And there’s only a VERY small PBM-contracting-side benefit to scale, so don’t think you’re going to magically convince Express Scripts to pay you more money for dispensing because you have 100 pharmacies. Unless you somehow have access to billions of dollars of capital to front your growth to 10,000 locations (how in the world is Ro’s valuation larger than Rite Aid? - also I think it would be a hilarious move if Ro offered to buy out RAD), the build out of a big internal PBM, and a massive sales team, I think that the advice that you need to “get big” is bad advice. (If you’re currently doing 15 prescriptions a day total, I’m not talking to you – it’s not 1941 anymore – you need more volume, but you already knew that).

I’d like to thank the NCPA team (and their predecessors at Lilly) for tirelessly publishing the Digest for nearly a century now. It’s an invaluable document, even if it isn’t perfect. To the pharmacist owners in my audience – when the call comes from NCPA to submit your data (in about may, usually) please step up and submit your data… as accurately and precisely as you can. Join NCPA if you haven’t. Trade associations like NCPA are a vital part of the antimonopoly tradition. Joining trade associations and actively participating in them is one way that you can support a future of regulated competition where we compete by improving the quality of products and services that we provide, instead of a world where we compete through financial engineering, price-fixing and idiotic games.