May 18, 2015 Food and Drug Administration Division of Dockets

May 18, 2015

Food and Drug Administration Division of Dockets Management (HFA-305) 5630 Fishers Lane, Room 1061 Rockville, MD 20852

Re: Proposed Rule to Require the Electronic Distribution of Prescribing Information for Human Prescription Drugs, Including Biological Products [Docket No. FDA–2007– N–0363]

Dear Sir or Madam:

The National Community Pharmacists Association (NCPA) represents the interests of America's community pharmacists, including the owners of nearly 23,000 independent community pharmacies. As small business owners and healthcare providers, our members have a vested interest in this proposed rule and appreciate the opportunity to submit our comments on this important topic.

The Food and Drug Administration (FDA) has formally proposed to amend its prescription drug and biological product labeling regulations to require electronic distribution of the prescribing information intended for health care professionals, which is currently distributed in paper form on or within the package from which the product is dispensed.

NCPA has a long history of working with industry stakeholders on this issue however, to date, a system whereby costs would not be shifted to community pharmacies has yet to be realized. As a point of reference, NCPA formally corresponded with both the FDA Commissioner and Director of the Office of Management and Budget (OMB) on April 4, 2014, respectfully requesting that OMB and FDA suspend plans to issue the pending proposed regulation “Electronic Distribution of Prescribing Information for Human Prescription Drugs Including Biological Products”, which was received by the Office of Information and Regulatory Affairs (OIRA) at OMB on August, 3, 2013.

NCPA Overall Concerns with the Proposed Rule

In the proposed rule, the FDA makes it clear that pharmacies will incur net costs due to initial capital costs to access the information, increased search time when accessing the information and the printing cost when a request is received for the information in printed form. The FDA estimates the annualized costs range from $47 million to $89 million and are a shift from manufacturers to pharmacies.

As small business owners, NCPA members are not in a position to absorb these costs. NCPA is opposed to the changes proposed in the rule until reliable systems are in place that will allow pharmacists to access electronic professional package inserts (PI) in a timely, efficient and cost neutral manner.

We would like to point to the July 2013 Government Accountability Office (GAO) report entitled Electronic Drug Labeling: No Consensus on the Advantages and Disadvantages of its Exclusive Use as support for our position. This report concludes that relying on electronic labeling as a complete substitute for paper labeling could adversely impact public health by limiting the availability of drug labeling for some physicians, pharmacists, and patients by requiring them to access drug labeling through a medium with which they might be uncomfortable, that they might find inconvenient, or that might be unavailable.1

NCPA’s primary recommendation for any move to electronic labeling is that costs not be shifted to small business community pharmacies. If paper drug labeling ceases to exist, costs will undoubtedly shift to the pharmacies to obtain and/or provide this information to patients who ask for it.

The GAO report noted that if patients want to continue receiving drug labeling in paper form and pharmacies are expected to print drug labeling for distribution, it would shift the costs of printing to the pharmacies.2 Please note that it is common practice in community pharmacies for pharmacists to utilize the manufacturer-provided paper insert, which patients often request.

A survey of pharmacists conducted recently found that pharmacists who prefer professional prescription labeling “indicate that these inserts are fast and easy to access, are familiar and allow pharmacists to readily search for needed information. Pharmacists also state that familiarity with the professional PIs allow them [to] access information reliably, with minimal interruption to work flow, and without errors.” Moreover, 27 percent of pharmacists polled indicated “that their pharmacy either does not have Internet access or that they cannot browse the internet.”3

Small business community pharmacies are unable to bear the costs of providing this information on their own, which would include additional computer terminals, printers and other office supplies such as paper, ink and toner. The GAO report also rightly noted that disruption to pharmacy workflow that would ensue from having to access the labeling electronically reduces the time available to counsel patients and has been shown to increase the risk for errors made when dispensing a drug.4

In addition, the FDA has not provided adequate evidence to support its assertion that the regulation would improve health benefits to patients. While the health benefits are unknown, the FDA cost benefit analysis shows minimal monetary benefits. In fact, the cost savings would merely be a financial shift from drug manufacturers to pharmacists.

Also, there is cause for concern that the proposed changes to the current labeling framework would lead to unintended public health negative consequences, including a risk that healthcare practitioners may not always have access to PI when it’s needed.

1United States Government Accountability Office Report to Congressional Committees; Electronic Drug Labeling No Consensus on the Advantages and Disadvantages of Its Exclusive Use; GAO-13-592; July 2013 2 Id. 3 NERA Economic Consulting, “Pharmacy Practice: A Report on Pharmacists’ Use of Printed Package Inserts” (Jan. 2015) at 7 (hereinafter, “NERA survey”). http://www.nera.com/publications/archive/2015/pharmacy-practice--a-report-on-pharmacists--use-of-printed-packa.html 4 Id. at 1.

NCPA Concerns with Shifting Financial Burden on Pharmacies and FDA’s Underestimation of Projected Costs

The proposed regulation would shift the financial burden from a higher gross margin industry (pharmaceutical manufacturing) to a lower gross margin industry (pharmacies), inevitably creating economic hardship for pharmacies.

The negative economic impacts of the proposed regulation affect pharmacies disproportionally. The FDA expects the cost effects would be similar between independent pharmacies and chain pharmacies (between $565 and $1,014 per affected pharmacy per year) and about 30% of expected costs to hospital pharmacies (between $1,436 and $2,684 per affected pharmacy per year).5 FDA’s assumption, however, is inaccurate. Since average annual revenues of independent pharmacies are less than chain and hospital pharmacies, additional annual costs as percentage of revenues are higher in independent pharmacies than other pharmacies.

Additionally, the FDA cost benefit analysis monetizes the time lost by pharmacists using the proposed system at between $31.9 million and $39.8 million annually6, but this estimate underestimates these costs because it fails to include time lost from using the automated phone line for prescribing information. Creating more work for pharmacists could exacerbate the issue of medication errors instead of reducing them.

NCPA is concerned that FDA’s Regulatory Impact Analysis7 regarding the Proposed Rule does not accurately weigh the rule’s proposed benefits against its likely costs. These issues are also addressed in detail in an independent, economic-focused assessment of FDA’s Regulatory Impact Analysis conducted by NDP Analytics (hereinafter “NDP Assessment”).8

According to the NDP Assessment, erroneous or unrealistic assumptions include the following:

Internet Access: The FDA cost benefit analysis assumes that all pharmacies including independent pharmacies have Internet access, but that some retail chains and hospitals partially or totally block employee access.9

The assumption is wrong. Inadequate or no Internet access is a reality in some parts of the United States. In its 2015 report, the FCC estimates that 54.6 million people and 17% of population in the U.S. in 2013 did not have access to the Internet.10. More than 32.6 million people in the rural and about 53% of population in the rural areas in 2013 did not have Internet access The FDA’s optimistic assumption underestimated the projected costs as well as unintended negative impacts on pharmacies and, by extension, patients.

5 Proposed Regulatory Impact Analysis, FDA, Table 17. 6 Proposed Regulatory Impact Analysis, FDA (pp. 26-27). 7 See “Electronic Distribution of Prescribing Information for Human Prescription Drugs, Including Biological Products: Proposed Regulatory Impact Analysis, Proposed Regulatory Flexibility Analysis, Unfunded Mandates Reform Act Analysis” available at: http://www.fda.gov/downloads/AboutFDA/ReportsManualsForms/Reports/EconomicAnalyses/UCM431225.pdf and FDA Docket FDA-2007-N-0363-0047. 8 See “Assessment of FDA’s Regulatory Impact Analysis Related to the FDA’s Proposed Regulation to Change the Prescription Information from Printed Materials to Electronic Version on the FDA Website” NDP Analytics (May 2015). This report is attached hereto as Attachment A. 9 Proposed Regulatory Impact Analysis, FDA (pp 25). 10 GAO’s report states: “Federal Communications Commission, approximately 14 million Americans have inadequate access or no access to adequate broadband capabilities.” GAO’s findings are based on a 2010 FCC report.

Computer Availability: The FDA assumes that only 1% of pharmacies would need to buy additional computers.11

This assumption is also wrong. Ho’s survey finds that the availability of computers ranged from 80% to 94% across different types of practice settings.12 Depending on the frequency of use and number of pharmacists, multiple computers may be needed. The FDA’s optimistic assumption underestimated the projected costs to pharmacies.

Referring patients to nearby pharmacy: In the emergency cases that cause limited access to electricity, Internet connectivity problems, and communication disruption, the FDA expects pharmacists would dispense limited drugs without using prescribing information, backup generators, or refer patients to another nearby pharmacy.13

The assumption is unrealistic, especially in the states such as Maine and Vermont where more than 60% of the population live in the rural communities. Data shows residents in certain rural communities had to travel an average 20 miles and up to 81 miles or 27 minutes and 88 minutes to find another pharmacy. 14 For patients needing timely fulfilment of prescription medications, it is unrealistic and inappropriate to assume ease of access to multiple pharmacies. The FDA’s unrealistic assumption underestimated the unintended negative consequences to patients.

Automated phone infrastructure: The FDA proposes manufacturers would be required to print a 24/7 toll-free telephone number in the statement appearing on the immediate container label and outer container or package that the healthcare professional could call to have the manufacturer send the most current prescribing information by fax, email, or mail.15

The assumption is unrealistic. The proposed alternative system that would require electronic prescribing information and ban printed prescribing information inserts would create negative health impacts, medication errors, and be inefficient. It is also unrealistic to assume that the phone line is an acceptable alternative to prescription information inserts, especially in emergency situations, because of the delay in receiving information. The FDA’s unrealistic assumption underestimated the costs to pharmacies and unintended consequences to patients.

NCPA Specific Concerns with FDA’s Proposed Solutions for a Complete Electronic System to Access Prescribing Information

NCPA appreciates FDA recognizing that in certain situations, Internet access will not be available to those seeking PI, and that as a solution FDA requires that the immediate container label and outer package bear a toll-free number for a health care professional to call to request the current PI by fax, email, or mail. This system could be used for those without regular Internet access or in the case of a

11 Proposed Regulatory Impact Analysis, FDA (pp 23-25). In the Federal Register, the FDA assumes 4% of all pharmacies did not have Internet access (pp 75512). 12 Ho, Yun-Xian, Quingxia Chen, Hui Nian, Kevin B Johnson. 2014. “An assessment of pharmacists’ readiness for paperless labeling: A National Survey.” Journal of the American Medical Informatics Association. 13 Proposed Regulatory Impact Analysis, FDA (pp 7-8). 14 Todd, Kelli, Fred Ulrich, and Keith Mueller. 2013. “Rural Pharmacy Closures: Implications for Rural Communities.” Rural Policy Brief, RUPRI Center for Rural Health Policy Analysis. 15 Federal Register, Vol. 79, No. 243, pp 75515.

public health emergency or natural disaster.16 NCPA believes this contingency plan is inadequate, particularly in emergency situations when fast back-up is needed most.

Those who access PI may prefer or need a printed version, especially for long, complex, detailed documents. Pharmacists may continue to rely on other printed resources at hand, such as official compendia, which are likely to be more outdated than the printed labeling that currently accompanies drug packaging. Alternatively, pharmacists may print PI from the FDA database to create their own “go-to” source of PI info. Whether healthcare professionals print PI or rely upon other resources, we have concerns that our members will not repeatedly visit the electronic database for drugs they use frequently.

We are also concerned that under a system that eliminates paper PI patient education may suffer. Pharmacists tout the importance of having a printed source of information to provide to patients, and to share with staff and/or other healthcare professionals.17 Consistent with this view, the GAO Report noted a concern among stakeholders that an electronic-only system would eliminate “the best method of ensuring that patients receive and understand drug labeling [which] is to hand it directly to them, accompanied with oral counseling.”18 NCPA is concerned that by in essence requiring pharmacists to print drug labeling for those patients who request it, pharmacists will opt to not provide the information to patients who request it, based on a disruption to workflow in addition to the costs associated with printing the lengthy document.

Conclusion

NCPA opposes the proposed rule at this time and would ask that it be withdrawn by FDA. We request that FDA continue requiring prescribing information in its current, printed format, on or within drug packaging.

Alternatively, NCPA could support a dual system, whereby prescribing information is available in both electronic and paper formats.

NCPA looks forward to working with the FDA to best meet their goals for healthcare providers using the most up-to-date prescribing information.

Sincerely,

Ronna B. Hauser, PharmD VP Pharmacy Affairs

Attachment A

16 See 79 Fed. Reg. at 75515. 17 See NERA Study, at 16. 18 See GAO Report at 13.

Assessment of the FDA’s Regulatory Impact Analysis Related to the FDA’s Proposed Regulation to Change the Prescription Information from Printed Materials to Electronic Version on the FDA Website

Nam D. Pham, Ph.D. David Stockwell, M.D. Mary Donovan

May 2015

Assessment of the FDA’s Regulatory Impact Analysis Related to the FDA’s Proposed Regulation to Change the Prescription Information from Printed Materials to Electronic Version on the FDA Website

Nam D. Pham, Ph.D., David Stockwell, M.D., and Mary Donovan

I. SUMMARY

The Food and Drug Administration (FDA) is proposing a regulation to replace (for the vast majority of prescription drugs) printed prescribing information, also referred to as the package insert, with an electronic version to be housed within the FDA website. The FDA cost benefit analysis estimates that the proposed regulation would produce between $5.0 million and $82.2 million net savings of operating costs annually.

The FDA, however, has not analyzed the health benefits or unintentional adverse effects on public health that may result from implementation of this rule. After reviewing the findings from the FDA cost benefit analysis, assumptions, related literature and statistics, we conclude that the proposed regulation does not provide economic or health benefits to Americans and is potentially counterproductive. The objective of the proposed rule is unsubstantiated and the proposed alternative system is inefficient. The FDA has already established several systems, including electronic prescribing information and drug recall, to communicate with healthcare providers, users of prescription information, and patients. While producing minimal economic benefits to the industry, the impact of the proposed regulation on public health is unknown. The FDA cost benefit analysis shows that the proposed rule will shift the financial burden from drug manufacturers to small pharmacies across the country in urban and rural areas alike. Furthermore, the proposed alternative system will produce unintended consequences of health risks to patients and economic hardship to pharmacies.

In short, a more thorough regulatory impact analysis would reveal that the FDA’s proposed regulation is unnecessary and unlikely to produce its purported benefits. The proposed regulation is likely to impose greater than anticipated negative costs and consequences. Users of prescribing information benefit from having information accessible in multiple channels. Therefore electronic information should be used to enhance rather than replace available printed package inserts.

Table 1 below summarizes the FDA estimates of costs and benefits. It also includes our qualitative assessment of areas that should have been included in the FDA’s estimates of costs and benefits, but that were not fully considered by the FDA. In failing to include these areas, the FDA has not accurately estimated social costs, benefits, and health impacts of the proposed regulation.

Table 1. Summary of Costs and Benefits

Costs and Benefits Amounts Net Annualized Economic Benefits Minimal; Less than FDA Estimates Benefits to drug manufacturers, repackers, & relabelers $51.8M ~ $170.8M Costs to pharmacists $46.4M ~ $88.7M Costs to blood product manufacturers $0.3M ~ $0.5M Net benefits to producers and users of package inserts $5.0M ~ $82.2M Costs to the FDA to administer/maintain proposed website Positive; No FDA estimates Health Impacts Uncertain Benefits to patients Non-consensus; No FDA estimates Potential risks to patients Many; No FDA estimates Negative Consequences Many Additional workload to pharmacists No FDA estimates Additional medication errors due to technology & human error No FDA estimates Economic hardship to pharmacies No FDA estimates

II. BACKGROUND

The FDA is proposing to amend its human prescription drug and biological product labeling regulations to (1) require electronic distribution of the prescribing information intended for healthcare professionals, and (2) eliminate, in large part, the distribution of printed prescribing information that accompanies a packaged drug product. Prescribing information provides health care professionals with the information necessary for the safe and effective use of the product. It is updated periodically to include the most current information, such as newly acquired safety information. Currently, the prescribing information is distributed in paper form on or within the package from which a prescription drug is dispensed.

The FDA has expressed concerns that package inserts may not contain the most current information because it may have been printed and distributed long before the time of the labeling changes. In order to ensure that the most current prescribing information for human prescription drugs and biological products can be updated in real time, the FDA is proposing the electronic form of prescribing information to be mandatory. The FDA is also proposing that prescribing information for most human prescription drugs and biological products will no longer be permitted to be distributed in paper form on or within the package.

Under the FDA proposed regulation, drug manufacturers would be required to submit the prescribing information for posting on the FDA’s publicly available labeling repository website (labels.fda.gov) and to update any change in the labeling. Manufacturers would also be required to replace the printed prescribing information currently available with an on-label statement directing users to the FDA’s labeling repository and a 24/7 toll-free telephone number (maintained by drug manufacturers) to receive requests for the manufacturer to email, fax, or mail paper copy of the prescribing information to users.

The FDA has conducted a cost benefit analysis to identify affected parties and to estimate the economic benefits of the proposed regulation. However, the FDA cost benefit analysis does not include an assessment of health benefits to American patients, distributional effects, and adverse effects. The FDA

analysis only focuses on the production cost savings for the prescribing information inserts. The FDA requests public comments on the rules as well as its cost benefit analysis.

III. OUR COMMENTS ON THE FDA REGULATORY IMPACT ANALYSIS

We reviewed the FDA regulatory impact analysis and its supporting documents. We also reviewed the existing research findings and statistics from governmental agencies and peer-reviewed professional journals to evaluate the FDA assumptions and expected economic and health impacts of the proposed rule. Our comments focus on three areas: the objectives of the FDA proposed regulation, the inefficiencies of the proposed rule, and the FDA cost benefit analysis, including discussion of the unintentional adverse effects of the FDA proposed regulation. None of these areas has received adequate consideration in the FDA’s analysis of the regulatory impacts of the proposed rule.

The Objective of the FDA Proposed Rule is Unsubstantiated

The FDA fails to provide evidence to support the need for a new regulation. The stated objective of the proposed rule is to ensure the most up to date prescribing information is available to health professionals in order to reduce errors caused by out of date information. Yet, evidence shows that medication errors due to package inserts are minimal. Furthermore, there is no evidence to support the FDA’s assertion that changing to an electronic system will improve public health.

There is no evidence that healthcare professionals are uninformed regarding key labeling changes

Under the current regulatory framework, there are already effective systems in place to communicate drug labeling changes to healthcare practitioners. The FDA has not provided any evidence to indicate that these systems are performing inadequately so as to warrant a significant change to current regulatory requirements.

According to the FDA internal review, an average of 500 safety labeling changes occurred each year during the period between 2003 and 2013. Labeling changes are classified by type and risk level. The boxed warning represents the highest alert and includes the most important safety information; contraindications include necessary advisory information for healthcare providers. Among the 500 safety label changes in each year, about 50 changes were related to box warnings and 60 changes were related to contraindications.1 Currently, there are between 150,000 and 200,000 unique SKUs of prescription drugs in the market and more than 4.0 billion prescription drugs were sold in the U.S. in 2014 (Table 2).2 Thankfully, less than 0.1% of products have significant safety label changes annually. However, when these changes do occur, it is important to issue notification to users immediately. The FDA currently has multiple channels in place to distribute drug alerts to healthcare practitioners and patients. For example, the FDA has an opt- in daily e-blast with drug updates and also maintains a system to recall drugs that might have severe health impacts on patients. Manufacturers are already required to submit the content of their current prescribing information in an electronic format to the FDA.3

1 Federal Register, Vol. 79, No. 243, pp 75506-75527. 2 “Total number of retail prescription filled at pharmacies,” State Health Facts, Kaiser Family Foundation http://kff.org/other/state- indicator/total-retail-rx-drugs/ 3 Proposed Regulatory Impact Analysis, FDA, pp 16.

Table 2. Average Annual Labeling Changes, Drug SKUs, and Retail Prescription Drugs4

Labeling changes per year 500 Box warnings 50 Contraindications 60 Number of SKUs 150,000~200,000 Number of retail prescription drugs filled in 2014 4,002,661,750

There is no evidence that package inserts are a significant contributor to medication errors

Despite the FDA’s concerns that package inserts may not be updated in a sufficiently timely manner and therefore might have created harm to patients, the FDA’s statistics show that less than 0.3% of medication errors are caused by these inserts. In fact, the FDA reported that 42.5% of medication errors are caused by human factors, leading by performance deficit (13.2%) and knowledge deficit (12.3%) (Table 3). Therefore, the proposed regulation to change the prescribing information inserts would have almost no impact on reducing medication errors.

Table 3. Medication Errors by Cause5

Causes % of Total Human Factors 42.5% Performance deficit 13.2% Knowledge deficit 12.3% Fatigue 0.3% Computer error 0.3% Labeling 19.9% Immediate container labels of product manufacturer 9.4% Labels of dispensed product 4.4% Carton labeling of product 4.4% Printed reference materials 0.9% Electronic reference materials 0.6% Package insert 0.3% Communications 18.8% Name Confusion 12.9% Packaging/Design 5.9%

Mistakes made by pharmacists are typically unrelated to erroneous or out-of-date package inserts. While overall there is a high rate of accuracy in medication dispensing (97.3%)6, mistakes made by pharmacists are contributing factors to medication errors and are found in different forms. Research findings over the past twenty years consistently showed that the leading cause of medication errors made by pharmacists is

4 Proposed Regulatory Impact Analysis, FDA; State Health Facts, Kaiser Family Foundation. 5 Thomas, Maria R., Carol Holquist, and Jerry Phillips. 2001. “Med error reports to FDA show a mixed bag.” FDA Safety Page, Drug Topics. 6 Flynn Elizabeth Ann, Kenneth N. Barker, and Brian J. Carnahan. 2003. “National Observational Study of Prescription Dispensing Accuracy and Safety in 50 Pharmacies.” Journal of the American Pharmaceutical Association.

their workload. For example, the Massachusetts Board of Registration in Pharmacy conducted a study to determine the impact of various factors on the incidence of medication errors made by practicing pharmacists in the 1990s. The study found that three leading causes to medication errors were too many telephone calls (62%), overload/unusual busy day (59%), and too many customers (53%).7 Similarly, a study that combined survey data from community pharmacies in 18 metropolitans in the mid-2000s also found a strong positive relationship between the risk of dispensing and pharmacy workload.8 Another national observational study of prescription dispensing accuracy and safety in 50 pharmacies across the U.S. found that errors in the computer order entry process used to create the label occur most frequently.9

As shown above, the proposed rule would not effectively reduce medication errors. Healthcare providers, physicians, and pharmacists do not rely solely on package inserts. Statistically, these inserts have not been the cause for medication errors. In fact, human factors including knowledge and heavy workloads with telephone calls and patients are leading causes to medication errors. In fact, many independent pharmacies in rural communities have only one pharmacist who performs multiple tasks with long hours throughout the year. 10 As the FDA described in its cost benefit analysis, the proposed alternative system would require pharmacists to increase their workload and to interact more with the Internet and telephone. As results, the proposed system will unintentionally create more medication errors.

The FDA presumption of health benefits is not supported by research findings

The FDA asserts that the proposed regulation will result in fewer prescribing errors due to out-of-date information.11 However, the FDA does not provide any research findings to support this assertion. In fact, research conducted by the Government Accountability Office (GAO) report found no consensus on the advantages and disadvantages of replacing printed prescribing information with an electronic-only substitute. Specifically the report states:

Some stakeholders we spoke with noted disadvantages that could offset any advantages gained from relying on electronic drug labeling as a complete substitute for paper labeling. Specifically, relying on electronic labeling could adversely impact public health by limiting the availability of drug labeling for some physicians, pharmacists, and patients by requiring them to access drug labeling through a medium with which they might be uncomfortable, they might find inconvenient, or that might be unavailable.12

In addition to the lack of evidence of package inserts contributing to medication errors, there is no consensus that the proposed rule would improve the status quo. Limiting the number of channels that healthcare professionals can use to access information only creates more challenges in providing accurate information and medication to patients.

7 Couris, R. Rebecca. 1999. “Medication Error Study.” Massachusetts Board of Registration in Pharmacy. 8 Malone, Daniel C., et al. 2007. “Pharmacist Workload and Pharmacy Characteristics Associated with the Dispensing of Potentially Clinically Important Drug-Drug Interactions.” Medical Care, Vol. 45, No. 5. 9 Flynn Elizabeth Ann, Kenneth N. Barker, and Brian J. Carnahan. 2003. “National Observational Study of Prescription Dispensing Accuracy and Safety in 50 Pharmacies.” Journal of the American Pharmaceutical Association. 10 Stratton, Timothy P. 2001. “The Economic Realities of Rural Pharmacy Practice.” The Journal of Rural Health. 11 Proposed Regulatory Impact Analysis, FDA, pp 5. 12 U.S. Government Accountability Office. 2013. “Electronic Drug Labeling: No Consensus on the Advantages and Disadvantages of Its Exclusive Use.” Web.

The Alternative System for Obtaining Prescribing Information is Inefficient and More Costly than Estimated

The proposed rule would require changing the product label on the immediate container label and any outer container or outside package to include a statement that directs users to the FDA’s labeling repository website to access prescribing information and a toll-free telephone number for requesting alternative options for obtaining the prescribing information via fax, email, or regular mail.

For pharmacies that lose access to the Internet due to environmental conditions, or operate in areas with unreliable or no access to the Internet, the FDA has proposed a solution that would require the availability of a 24/7 phone line that could be used to request prescribing information. However, this solution would add multiple steps to the process of accessing prescribing information. After placing the phone call, the pharmacist must wait to receive the information from the manufacturer. It is reasonable to assume that, if the pharmacist had to contact the manufacturer for information because it could not be accessed online, the information also cannot be received via email. After placing the request on the phone, the pharmacist must wait to receive the information via regular mail or fax. In contrast, the package inserts are currently included with the medication upon delivery of the product, so a pharmacist has immediate access if needed, without any extra steps or delay.

Maximizing efficiency and minimizing errors in dispensing medication at the pharmacy level requires low search cost for accessing information and a work environment that minimizes errors. Because pharmacists currently have the option to use a variety of resources to access prescribing information, including the package insert, electronic and paper compendia, and DailyMed, they are able to obtain prescribing information in a format that is most comfortable and easy to access. Additionally, the information provided in every package of prescription drugs allows health professionals to have access to important prescribing information regardless of the circumstances. In fact, a 2015 study conducted by NERA, found that 80% of pharmacists surveyed indicated that they checked the package insert at least once a month.13

Additionally, the FDA cost benefit analysis monetizes the time lost by pharmacists using the proposed system at between $31.9 million and $39.8 million annually14, but this estimate underestimates these costs because it fails to include time lost from using the automated phone line for prescribing information. As discussed in detail earlier, a pharmacist being overworked is a leading contributing factor to medication errors at the pharmacy level. As a result, creating more work for pharmacists would exacerbate the issue of medication errors instead of reducing them.

The FDA Cost-Benefit Analysis is Incomplete and Shows that Users and Small Businesses are Disproportionally Affected by the Proposed Rule

While the FDA’s cost benefit analysis shows a net benefit, in reality, there is a transfer of costs and no substantive evidence supporting an improvement in public health. Additionally, we found several assumptions in the analysis to be unrealistic or erroneous and therefore directly affect the results of the

13 Butler, Sarah. 2015. “Pharmacy Practice: A Report on Pharmacists’ Use of Printed Packaged Inserts.” NERA Economic Consulting. Web. 14 Proposed Regulatory Impact Analysis, FDA, pp. 26-27.

cost benefit analysis as well as the distributional effects of the proposed regulation. First, using the costs and benefits derived from the FDA’s analysis, we show that the proposed regulation does not improve social welfare and, in fact, simply shifts the financial burden from drug manufacturers to tens of thousands small pharmacies. The FDA overlooks the unintended economic hardship to small businesses, which will spillover to patients, especially in the rural communities where pharmacists are already overworked to provide their multiple services. Second, we discuss the incorrect assumptions that the FDA relies upon in its regulatory analysis.

While net cost savings to drug manufacturers are immaterial, a significant financial burden is shifted to users of prescribing information

The annual net cost savings would be generated from savings to drug manufacturers, repackers, and relabelers at the expense of retail pharmacies and blood product manufacturers. The FDA estimates that net cost savings of the regulation would be between $5.0 million and $82.2 million per year.15

The FDA estimates that the one-time administrative and production costs to drug manufacturers, repackers, and relabelers would be between $191.4 million and $326.7 million. Annual recurring costs to drug manufacturers, repackers, and relabelers, including nonstandard labels and toll-free numbers, are estimated between $5.5 million and $20.5 million. The FDA estimates that drug manufacturers, repackers, and relabelers would save between $93.8 million and $216.6 million per year on storage and printing costs of prescribing information on paper. After monetizing one-time costs over 10 years, the FDA estimates drug manufacturers, repackers, and relabelers would save between $51.8 million and $170.8 million per year (Table 4).

On the other hand, the proposed rule creates new costs for the users of prescribing information. The FDA cost benefit analysis assumes pharmacists continue to use prescribing information as a key source of drug product information and physicians’ methods of accessing this information would not change. The FDA estimates one-time costs to pharmacies, including computer purchases and staff training, range between $7.2 million and $18.6 million; annual recurring costs to pharmacies for staff time lost and printing costs would be between $45.4 million and $86.0 million. After monetizing one-time costs over 10 years, expected annual costs to pharmacies are between $46.4 million and $88.7 million per year. In addition to pharmacies, the FDA estimated the proposed regulation would generate between $0.3 and $0.5 million annual costs to blood product manufactures. Subtracting the costs imposed on these entities, from the estimated savings attributed to manufacturers, repackers, and relabelers, the FDA estimated an annualized net savings between $5.0 million and $82.2 million (Table 4).

Notably, however, none of the net cost savings are actually enjoyed by the users of prescribing information. Further, the estimated cost savings is immaterial to the prescription drug industry which is estimated to have generated $305.8 billion of revenue in 2014.16

15 Proposed Regulatory Impact Analysis, FDA, Table 19. 16 Fein, Adam J. 2015. “2014-15 Economic Report on Retail, Mail, and Specialty Pharmacies.” Drug Channels Institute.

Table 4. Estimated Annual Costs and Benefits17

Stakeholders Monetary Impacts Annualized benefits to drug manufacturers, repackers, & relabelers $51.8M ~ $170.8M One-time costs $191.4M ~ $326.7M Annual costs $5.5M ~ $20.5M Annual savings $93.8M ~ $216.6M Annualized costs to pharmacies $46.4M ~ $88.7M One-time costs $7.2M ~ $18.6M Annual costs $45.4M ~ $ 86.0M Annualized costs to blood product manufacturers $0.3M ~ $0.5M Annualized net savings to producers and users $5.0M ~ $82.2M Estimated 2014 Prescription Revenue 18 $305.8 Billion

The shifted financial burden on pharmacies outweighs the cost savings to drug manufacturers

Although the net cost savings are immaterial to the prescription drug industry, the shifted cost burden would have great financial impacts on pharmacies. During the period between 2006 and 2013, cost of goods sold (purchases of drugs) accounted for 76.3% of total sales of pharmacies and drugstores. Total operating expenses, including payroll and taxes, accounted for an average 22.7% of total sales of pharmacies and drugstores. Gross margin (sales minus cost of goods sold) of pharmacies and drugstores averaged 23.7% of total sales and net profit margin (gross margin minus total operating expenses) averaged 1.0% of total sales (Table 5). For comparison purposes, gross margin in pharmacies and drugstores is lower than in all retailer industries (27.3%), all manufacturing sectors (40.9%), and pharmaceutical manufacturing industry (69.6%).

The proposed regulation would shift the financial burden from a higher gross margin industry to a lower gross margin industry, inevitably creating economic hardship for pharmacies and drugstores. Moreover, these pharmacies are small businesses and would be more vulnerable to additional expenditures. Based on the U.S. Census data, about 36% of total pharmacies have fewer than 9 employees and 64% have fewer than 19 employees (Table A.1 in the Appendix provides pharmacies by employment size by state).19

Table 5. Gross and Net Margins of Pharmacies and Drugstores, 2006-201320

Average 2006-2013 Sales 100.0% Cost of Goods Sold 76.3% Gross Margin 23.7% Total Operating Expenses (incl. taxes) 22.7% Net Profits 1.0%

17 Proposed Regulatory Impact Analysis, FDA. 18 Fein, Adam J. 2015. “2014-15 Economic Report on Retail, Mail, and Specialty Pharmacies.” Drug Channels Institute. 19 U.S. Census Bureau, American FactFinder. 20 U.S. Census Bureau, Monthly and Annual Retail Trade; http://www.census.gov/retail/index.html

The FDA underestimates the number of affected parties and regulatory costs

The FDA concluded that the proposed rule would have significant negative economic impacts on a substantial number of small business entities. The FDA estimated the benefits of between $51.8 million and $170.8 million per year will go to between 551 and 700 prescription drug manufacturers and between 900 and 1,300 repackers and relabelers. In contrast, 79,206 pharmacies and 118 licensed blood product manufacturers will share the costs of between $46.7 million and $89.2 million per year. The FDA expects physicians and nurses to have minimum effects from the proposed regulation. Although acknowledging that thin paper manufacturers and printing companies will be affected negatively, the FDA has not included costs to these stakeholders in its overall cost benefit analysis. In addition, the FDA analysis does not mention additional costs to the FDA to maintain the website to manage the prescribing information. We expect the FDA would need one-time costs and annual recurring costs to upgrade its system for additional visitors, cybersecurity protection, redundancy, and uptime. The FDA would also need additional resources to administer, review, and process the prescription information in a timely manner. Thus, the net cost savings would be lower than the current FDA estimates in the cost benefit analysis (Table 6).

Table 6. Affected Parties -- Positive and Negative21

Number of Entities Short- and Long-term Effects Positive Effects Drug manufacturers, 551~700 drug Short-term costs but long-term repackers, & relabelers; manufacturers benefits distributors & wholesalers of 900~1,300 repackers & prescription drugs relabelers Negative Effects Pharmacies 79,206 Both short- and long-term costs Blood product manufacturers 118 licensed manufacturers Both short- and long-term costs Physicians 337,213 Minimal impacts; No FDA estimates Thin paper manufacturers & 3 major manufacturers and Both short- and long-term costs; printing companies few smaller manufacturers No FDA estimates FDA Both short- and long-term costs; No FDA estimates

In addition to the parties that the FDA included in the regulatory impact analysis, patients will also be affected directly by the proposed regulation. Unlike the assumption by the FDA that the prescribing inserts are only used by healthcare professionals and not patients, the GAO report found that some pharmacists use printed inserts to counsel patients. The GAO report also states that pharmacists’ workflow will be disrupted by searching on the Internet and/or printing the inserts.22 The use of the insert to inform the patient is further supported by the NERA study, which found that 55% of pharmacists surveyed have

21 Proposed Regulatory Impact Analysis, FDA. 22 U.S. Government Accountability Office. 2013. “Electronic Drug Labeling: No Consensus on the Advantages and Disadvantages of Its Exclusive Use.” Web, pp 6 and 13.

provided the printed insert to a patient.23 Overall, the proposed regulation to ban printed insert materials will directly affect patients via multiple channels. In the event of the Internet is not available, pharmacists will not have access to show patients the printed materials. The amount of time that pharmacists would need to search on the Internet and print the material inserts would reduce the time that pharmacists can spend with patients.

The FDA proposed rule has unintended consequences that negatively affect small businesses and rural pharmacies

Because the proposed rule shifts the cost burden from drug manufacturers to pharmacies, there are unintended consequences to small businesses across the U.S., especially those in rural areas. Independent pharmacies serve their communities in ways far beyond dispensing prescription drugs. The National Community Pharmacy Association notes that independent pharmacies “employ over 220,000 full- time equivalent workers, helping to stimulate local economies, paying state and local taxes, and providing high quality services that make a difference in the daily lives of patients.”24

The proposed regulation hurts small independent pharmacies more than other chain and hospital pharmacies

The negative economic impacts of the proposed regulation affect pharmacies disproportionally. While registering the highest number of establishments across the country, independent pharmacies accounted for less than 15% of total dollar sales of prescription drugs in 2013. Since independent pharmacies often serve a smaller population, their annual revenues are also lower than other types of pharmacies. About 33% of independent pharmacies had sales under $2.5 million a year, 53% of independent pharmacies had sales between $2.5 million and $6.5 million per year, and only 14% of independent pharmacies had sales above $6.5 million per year.25

The FDA expects the cost effects would be similar between independent pharmacies and chain pharmacies (between $565 and $1,014 per affected pharmacy per year) and about 30% of expected costs to hospital pharmacies (between $1,436 and $2,684 per affected pharmacy per year).26 The FDA’s assumption, however, is inaccurate. Since average annual revenues of independent pharmacies are less than chain and hospital pharmacies, additional annual costs as percentage of revenues are higher in independent pharmacies than other pharmacies.

The proposed regulation would unintentionally hurt patients in rural communities

Studies have shown that the financial performance of independent pharmacies has been declining over the years and has affected rural communities negatively.27 With additional financial burdens and restrictive regulations, independent pharmacies will face difficulties to survive. The negative economic impacts of the

23 Butler, Sarah. 2015. “Pharmacy Practice: A Report on Pharmacists’ Use of Printed Packaged Inserts.” NERA Economic Consulting. 24 2014 NCPA Digest, National Community Pharmacists Association. 25 2014 NCPA Digest, National Community Pharmacists Association. 26 Proposed Regulatory Impact Analysis, FDA, Table 17. 27 Todd, Kelli, Fred Ulrich, and Keith Mueller. 2013. “Rural Pharmacy Closures: Implications for Rural Communities.” Rural Policy Brief, RUPRI Center for Rural Health Policy Analysis.

proposed regulation on rural independent pharmacies would spillover to patients in those communities. Independent pharmacies are providing vital services to rural areas. Nearly 22,000 pharmacies out of total 60,221 pharmacies (36.5% of total) in the U.S. are independent and owned by individuals. Approximately 73% of independent pharmacies are located in areas with a population of fewer than 50,000 persons; half of these are located in the areas with fewer than 10,000 persons. The role of independent pharmacies is highly essential in states where independent pharmacies account for a large share of total pharmacies. For example, independent pharmacies account for 79% of total pharmacies in North Dakota, 58% in South Dakota, 57% in Arkansas, and 56% in Oklahoma (Table 7) (Table A.2 in the Appendix provides distribution of pharmacies by type and by state).28

Table 7. Distribution of Pharmacies in the United States, by type, 201329

Independent Chain Supermarket Mass Merchant United States 36.5% 35.3% 13.8% 14.5% North Dakota 79.3% 20.7% 0.0% 0.0% South Dakota 58.4% 15.6% 7.5% 18.5% Arkansas 56.8% 9.7% 9.7% 23.7% Oklahoma 56.5% 22.4% 4.5% 16.5% Montana 53.2% 12.0% 18.1% 16.7%

The unrealistic and erroneous assumptions in the FDA cost benefit analysis cause the FDA to underestimate the overall regulatory costs of the proposed rule

A combination of the optimism in the assumptions and the incompleteness of the analysis underestimated the social costs and overestimated the benefits of proposed regulation. Furthermore, the cost benefit analysis left out the crucial health effects (both positive and negative) and other unintended negative effects to Americans across the country. As acknowledged by the FDA, because of the large number of companies impacted, small changes in assumptions can create meaningful differences in outcomes. Our assessment of erroneous or unrealistic assumptions is below.

The assumption is wrong. Inadequate or no Internet access is a reality in some parts of the United States. In its 2015 report, the FCC estimates that 54.6 million people and 17% of population in the U.S. in 2013 did not have access to the Internet.31. More than 32.6 million people in the rural and about 53% of population in the rural areas in 2013 did not have Internet access (Table A.3 in the Appendix provides population without Internet access in all areas and rural areas by state).32 Furthermore, IT disruptions at the FDA as well as

28 2014 NCPA Digest, National Community Pharmacists Association. 29 2014 NCPA Digest, National Community Pharmacists Association. 30 Proposed Regulatory Impact Analysis, FDA, pp 25. 31 GAO’s report states: “Federal Communications Commission, approximately 14 million Americans have inadequate access or no access to adequate broadband capabilities.” GAO’s findings are based on a 2010 FCC report. 32 FCC. 2015. “2015 Broadband Progress Report and Notice of Inquiry of Immediate Action to Accelerate Deployment.” Before the Federal Communications Commission.

third-party Internet service providers have not been considered. According to two recent surveys in June 2014 conducted by Symantec, 70% of government agencies experienced downtime of 30 minutes or more in a month. The surveys found that network or server outage and Internet connectivity loss are top two causes of downtime, 42% and 29%, respectively. Nine out of 10 government field workers, who participated in the surveys, said their agency’s most recent downtime affected their ability to do their job.33 All told, the FDA’s optimistic assumption underestimated the projected costs as well as unintended negative impacts on pharmacies and, by extension, patients.

Computer Availability: The FDA assumes that only 1% of pharmacies would need to buy additional computers.34

This assumption is also wrong. Ho’s survey finds that the availability of computers ranged from 80% to 94% across different types of practice settings.35 Depending on the frequency of use and number of pharmacists, multiple computers may be needed. The FDA’s optimistic assumption underestimated the projected costs to pharmacies.

The assumption is unrealistic, especially in the states such as Maine and Vermont where more than 60% of the population live in the rural communities. Data shows residents in certain rural communities had to travel an average 20 miles and up to 81 miles or 27 minutes and 88 minutes to find another pharmacy. 37 For patients needing timely fulfilment of prescription medications, it is unrealistic and inappropriate to assume ease of access to multiple pharmacies. Pharmacies operating without access to prescribing information may result in increased errors. The FDA’s unrealistic assumption underestimated the unintended negative consequences to patients.

Automated phone infrastructure: The FDA proposes manufacturers would be required to print a 24/7 toll- free telephone number in the statement appearing on the immediate container label and outer container or package that the healthcare professional could call to have the manufacturer send the most current prescribing information by fax, email, or mail.38

33 Hunter, Lindsey. 2014. “The Drive to Thrive: Ensuring the Agile Data Center.” Meritalk, Underwritten by Symantec. 34 Proposed Regulatory Impact Analysis, FDA, pp 23-25. In the Federal Register, the FDA assumes 4% of all pharmacies did not have Internet access, pp 75512. 35 Ho, Yun-Xian, Quingxia Chen, Hui Nian, and Kevin B Johnson. 2014. “An assessment of pharmacists’ readiness for paperless labeling: A National Survey.” Journal of the American Medical Informatics Association. 36 Proposed Regulatory Impact Analysis, FDA, pp 7-8. 37 Todd, Kelli, Fred Ulrich, and Keith Mueller. 2013. “Rural Pharmacy Closures: Implications for Rural Communities.” Rural Policy Brief, RUPRI Center for Rural Health Policy Analysis. 38 Federal Register, Vol. 79, No. 243, pp 75515.

alternative to prescription information inserts, especially in emergency situations, because of the delay in receiving information. The FDA’s unrealistic assumption underestimated the costs to pharmacies and unintended consequences to patients.

IV. CONCLUSION

The FDA’s proposed regulation to replace printed prescribing information materials with an electronic-only version would shift the production costs from drug manufacturers to pharmacies. The FDA has not provided adequate evidence to support its assertion that the regulation would improve health benefits to patients. While the health benefits are unknown, the FDA cost benefit analysis shows minimal monetary benefits. In fact, the cost savings would merely be a financial shift from drug manufacturers to pharmacists. Furthermore, the FDA cost benefit analysis used unrealistic assumptions that underestimate the costs to users of prescribing information and patients. The FDA’s proposed regulation will increase the workload for all pharmacists and create economic hardship for small pharmacies that will have negative health impacts on patients across the country in the urban and rural communities alike. By banning the prescribing information inserts that are currently available, the FDA cuts back important information to assist pharmacists and healthcare providers to do their work properly. Our analysis suggests the FDA should continue to require prescribing information in its current, printed format, on or within drug packaging. The FDA should consider the development of an electronic labeling repository only in addition to the current labeling system.

References

Butler, Sarah. 2015. “Pharmacy Practice: A Report on Pharmacists’ Use of Printed Packaged Inserts.” NERA Economic Consulting.

Couris, R. Rebecca. 1999. “Medication Error Study.” Massachusetts Board of Registration in Pharmacy.

Fein, Adam J. 2015. “2014-15 Economic Report on Retail, Mail, and Specialty Pharmacies.” Drug Channels Institute.

Flynn Elizabeth Ann, Kenneth N. Barker, and Brian J. Carnahan. 2003. “National Observational Study of Prescription Dispensing Accuracy and Safety in 50 Pharmacies.” Journal of the American Pharmaceutical Association.

Ho, Yun-Xian, Quingxia Chen, Hui Nian, and Kevin B Johnson. 2014. “An assessment of pharmacists’ readiness for paperless labeling: A National Survey.” Journal of the American Medical Informatics Association.

Hunter, Lindsey. 2014. “The Drive to Thrive: Ensuring the Agile Data Center.” Meritalk, Underwritten by Symantec.

Malone, Daniel C., Jacob Abarca, Grant Skrepnek, John E. Murphy, Edward P. Armstrong, Amy J. Grizzle, Rick A. Rehfeld, and Raymond L. Woosley. 2007. “Pharmacist Workload and Pharmacy Characteristics Associated with the Dispensing of Potentially Clinically Important Drug-Drug Interactions.” Medical Care, Vol. 45, No. 5.

National Community Pharmacists Association. 2014. “2014 NCPA Digest.” National Community Pharmacists Association.

State Health Facts. 2015. Kaiser Family Foundation.

Stratton, Timothy P. 2001. “The Economic Realities of Rural Pharmacy Practice.” The Journal of Rural Health.

Thomas, Maria R., Carol Holquist, and Jerry Phillips. 2001. “Med error reports to FDA show a mixed bag.” FDA Safety Page, Drug Topics.

Todd, Kelli, Fred Ulrich, and Keith Mueller. 2013. “Rural Pharmacy Closures: Implications for Rural Communities.” Rural Policy Brief, RUPRI Center for Rural Health Policy Analysis.

U.S. Census Bureau. 2015. American FactFinder.

U.S. Census Bureau. 2015. Monthly and Annual Retail Trade.

U.S. Federal Communications Commission. 2015. “2015 Broadband Progress Report and Notice of Inquiry of Immediate Action to Accelerate Deployment.” Before the Federal Communications Commission.

U.S. Federal Register. Vol. 79, No. 243, pp 75506-75527.

U.S. Food & Drug Administration. Proposed Regulatory Impact Analysis, FDA.

U.S. Government Accountability Office. 2013. “Electronic Drug Labeling: No Consensus on the Advantages and Disadvantages of Its Exclusive Use.”

Appendix

Table A.1: Pharmacy by Employment Size and by State, 201239

Establishments with Establishments with Establishments with State 1-9 employees 10-19 employees >20 employees United States 36% 28% 36% Alabama 42% 34% 24% Alaska 49% 21% 31% Arizona 20% 24% 56% Arkansas 55% 28% 18% California 38% 23% 39% Colorado 37% 24% 39% Connecticut 21% 28% 51% Delaware 16% 32% 52% District of Columbia 25% 29% 45% Florida 39% 20% 42% Georgia 35% 39% 26% Hawaii 47% 15% 38% Idaho 37% 38% 25% Illinois 32% 19% 49% Indiana 25% 25% 50% Iowa 44% 23% 33% Kansas 43% 26% 30% Kentucky 37% 39% 24% Louisiana 42% 29% 29% Maine 27% 53% 20% Maryland 34% 30% 36% Massachusetts 15% 27% 58% Michigan 38% 36% 26% Minnesota 36% 23% 41% Mississippi 50% 32% 18% Missouri 35% 28% 38% Montana 53% 33% 14% Nebraska 43% 23% 34% Nevada 22% 30% 49% New Hampshire 14% 38% 48% New Jersey 36% 27% 36% New Mexico 33% 21% 46% New York 46% 21% 32% North Carolina 26% 40% 33% North Dakota 57% 30% 13% Ohio 22% 38% 40% Oklahoma 46% 24% 29% Oregon 33% 32% 35% Pennsylvania 32% 34% 34%

39 U.S. Census Bureau, American FactFinder.

Rhode Island 16% 32% 51% South Carolina 30% 38% 33% South Dakota 56% 18% 27% Tennessee 36% 31% 33% Texas 43% 21% 36% Utah 45% 27% 29% Vermont 28% 46% 25% Virginia 25% 35% 41% Washington 30% 31% 39% West Virginia 32% 47% 21% Wisconsin 34% 18% 48% Wyoming 52% 24% 24%

Table A.2: Pharmacy by Type and by State, 201340

Panel A. Number of Pharmacies

Mass State Independent Chain Supermarket Total Merchant United States 21,966 21,238 8,301 8,716 60,221 Alabama 589 374 116 183 1,262 Alaska 33 9 25 32 99 Arizona 142 394 253 177 966 Arkansas 402 69 69 168 708 California 2,093 2,109 639 593 5,434 Colorado 153 185 255 163 756 Connecticut 151 308 91 61 611 Delaware 29 109 17 18 173 District of Columbia 42 68 22 3 135 Florida 1,352 1,602 948 514 4,416 Georgia 729 754 426 729 2,638 Hawaii 79 61 10 30 180 Idaho 117 44 44 54 259 Illinois 689 993 169 354 2,205 Indiana 205 517 171 207 1,100 Iowa 309 216 26 120 671 Kansas 271 148 73 111 603 Kentucky 524 299 118 132 1,073 Louisiana 507 326 92 167 1,092 Maine 65 114 68 22 269 Maryland 355 381 217 126 1,079 Massachusetts 204 624 124 93 1,045 Michigan 953 778 153 374 2,258 Minnesota 309 239 127 192 867 Mississippi 374 156 49 188 767 Missouri 503 306 146 197 1,152 Montana 115 26 39 36 216 Nebraska 214 116 27 72 429 Nevada 81 204 123 90 498 New Hampshire 37 132 39 38 246 New Jersey 749 748 298 141 1,936 New Mexico 89 93 46 65 293 New York 2,294 1,650 281 256 4,481 North Carolina 699 830 176 273 1,978 North Dakota 119 31 - - 150 Ohio 532 881 336 373 2,122 Oklahoma 461 183 37 135 816 Oregon 137 134 131 185 587

40 2014 NCPA Digest, National Community Pharmacists Association.

Pennsylvania 994 1,126 369 278 2,767 Rhode Island 28 125 15 15 183 South Carolina 352 435 169 153 1,109 South Dakota 101 27 13 32 173 Tennessee 569 426 231 137 1,363 Texas 1,607 1,351 751 628 4,337 Utah 183 67 64 87 401 Vermont 44 58 19 7 128 Virginia 394 571 309 234 1,508 Washington 343 326 234 196 1,099 West Virginia 226 189 58 72 545 Wisconsin 374 307 59 178 918 Wyoming 45 19 29 27 120

Panel B. As % of Total Pharmacies

Mass State Independent Chain Supermarket Total Merchant United States 36% 35% 14% 14% 100% Alabama 47% 30% 9% 15% 100% Alaska 33% 9% 25% 32% 100% Arizona 15% 41% 26% 18% 100% Arkansas 57% 10% 10% 24% 100% California 39% 39% 12% 11% 100% Colorado 20% 24% 34% 22% 100% Connecticut 25% 50% 15% 10% 100% Delaware 17% 63% 10% 10% 100% District of Columbia 31% 50% 16% 2% 100% Florida 31% 36% 21% 12% 100% Georgia 28% 29% 16% 28% 100% Hawaii 44% 34% 6% 17% 100% Idaho 45% 17% 17% 21% 100% Illinois 31% 45% 8% 16% 100% Indiana 19% 47% 16% 19% 100% Iowa 46% 32% 4% 18% 100% Kansas 45% 25% 12% 18% 100% Kentucky 49% 28% 11% 12% 100% Louisiana 46% 30% 8% 15% 100% Maine 24% 42% 25% 8% 100% Maryland 33% 35% 20% 12% 100% Massachusetts 20% 60% 12% 9% 100% Michigan 42% 34% 7% 17% 100% Minnesota 36% 28% 15% 22% 100% Mississippi 49% 20% 6% 25% 100% Missouri 44% 27% 13% 17% 100% Montana 53% 12% 18% 17% 100% Nebraska 50% 27% 6% 17% 100% Nevada 16% 41% 25% 18% 100%

New Hampshire 15% 54% 16% 15% 100% New Jersey 39% 39% 15% 7% 100% New Mexico 30% 32% 16% 22% 100% New York 51% 37% 6% 6% 100% North Carolina 35% 42% 9% 14% 100% North Dakota 79% 21% 0% 0% 100% Ohio 25% 42% 16% 18% 100% Oklahoma 56% 22% 5% 17% 100% Oregon 23% 23% 22% 32% 100% Pennsylvania 36% 41% 13% 10% 100% Rhode Island 15% 68% 8% 8% 100% South Carolina 32% 39% 15% 14% 100% South Dakota 58% 16% 8% 18% 100% Tennessee 42% 31% 17% 10% 100% Texas 37% 31% 17% 14% 100% Utah 46% 17% 16% 22% 100% Vermont 34% 45% 15% 5% 100% Virginia 26% 38% 20% 16% 100% Washington 31% 30% 21% 18% 100% West Virginia 41% 35% 11% 13% 100% Wisconsin 41% 33% 6% 19% 100% Wyoming 38% 16% 24% 23% 100%

Table A.3: Internet Access in Urban and Rural Areas by State, 201341

Urban Rural State Population Without Percentage of Population Without Percentage of Access (1,000s) Population Access (1,000s) Population United States 54,560.0 17% 32,628.3 53% Alabama 1,701.3 35% 1,113.2 56% Alaska 284.8 38% 198.8 81% Arizona 1,162.2 17% 540.1 80% Arkansas 1,751.5 59% 1,083.8 84% California 2,601.4 7% 1,266.4 67% Colorado 942.8 18% 504.5 71% Connecticut 49.9 1% 20.6 5% Delaware 29.7 3% 20.1 13% District of Columbia 9.4 2% 0.0 0% Florida 1,278.3 7% 707.7 41% Georgia 1,403.1 14% 1,100.5 44% Hawaii 57.0 4% 49.6 45% Idaho 820.3 50% 374.0 79% Illinois 709.6 5% 570.2 39% Indiana 947.1 14% 786.7 44% Iowa 761.3 25% 651.9 60% Kansas 793.9 27% 512.3 71% Kentucky 1,766.7 40% 1,322.1 73% Louisiana 1,324.8 29% 793.5 64% Maine 290.7 22% 257.9 31% Maryland 417.5 7% 201.5 27% Massachusetts 238.0 4% 98.3 18% Michigan 1,250.1 13% 968.5 39% Minnesota 724.7 13% 669.4 47% Mississippi 1,196.5 40% 935.0 61% Missouri 1,791.5 29% 1,268.6 71% Montana 882.3 87% 399.9 90% Nebraska 502.2 27% 348.6 73% Nevada 173.2 6% 98.4 64% New Hampshire 227.6 17% 191.0 36% New Jersey 172.3 2% 60.3 13% New Mexico 643.0 30% 356.2 77% New York 518.0 3% 496.4 21% North Carolina 1,446.2 14% 1,153.7 35% North Dakota 102.3 15% 97.9 37% Ohio 1,934.8 17% 1,229.8 49% Oklahoma 1,882.3 49% 1,144.6 89% Oregon 276.3 7% 247.7 34% Pennsylvania 1,695.4 13% 978.6 36%

41 FCC. 2015. “2015 Broadband Progress Report and Notice of Inquiry of Immediate Action to Accelerate Deployment.” Before the Federal Communications Commission.

Rhode Island 6.6 1% 6.4 6% South Carolina 1,085.5 23% 698.0 44% South Dakota 160.3 19% 157.4 45% Tennessee 1,148.4 18% 988.9 45% Texas 9,986.8 38% 3,334.1 84% Utah 149.2 5% 104.9 39% Vermont 502.4 80% 345.9 90% Virginia 1,733.6 21% 1,280.5 64% Washington 276.7 4% 237.6 21% West Virginia 1,041.9 56% 713.6 74% Wisconsin 961.7 17% 909.9 53% Wyoming 175.2 30% 152.3 74%

About the Authors

Nam D. Pham, PhD Managing Partner, ndp | analytics

Nam D. Pham is Managing Partner of ndp | analytics, a strategic research firm that specializes in economic analysis of public policy and legal issues. Clients of ndp | analytics include trade associations, financial institutions, law firms, U.S. and foreign corporations, and multinational organizations. Prior to founding ndp | analytics in 2000, Dr. Pham was Vice President at Scudder Kemper Investments in Boston, where he was responsible for research, asset allocations, and currency hedging for global and international bond funds. Before that he was Chief Economist of the Asia Region for Standard & Poor’s DRI; an economist at the World Bank; and a consultant to both the Department of Commerce and the Federal Trade Commission. Dr. Pham is an adjunct professor at the George Washington University. Dr. Pham holds a Ph.D. in economics from the George Washington University, an M.A. from Georgetown University; and a B.A. from the University of Maryland. He is a member of the board of advisors to the Dingman Center for Entrepreneurship at the University of Maryland School of Business and a board member of the Food Recovery Network.

David Stockwell, MD Senior Advisor, ndp | analytics Associate Professor of Pediatrics, The George Washington University School of Medicine Vice President, Clinical Services, Pascal Metrics-a Patient Safety Organization

Dr. Stockwell is a nationally recognized quality improvement expert. He recently served as the Executive Director of Improvement Science as well as the Medical Director of the Pediatric Intensive Care Unit at Children’s National Medical Center. For over 15 years, Dr. Stockwell has been a driving force in the performance improvement of clinical teams. He leads a multi-hospital collaborative that investigates automated adverse event detection by using electronic medical records and is partnering with the FDA to develop tools quantifying the impact of adverse events in children. Dr. Stockwell also co-leads an AHRQ funded national research effort to develop a state of the art pediatric patient safety measurement system. Dr. Stockwell has been named one of the “Best Doctors in America” continuously since 2007. He received his Doctor of Medicine degree from the University of Oklahoma College of Medicine, Master’s in Business Administration from The George Washington University, and a Bachelor’s degree from University of Oklahoma.

Mary Donovan Senior Associate, ndp | analytics

As a Senior Associate at ndp | analytics, Mary serves dual roles of economist and communications manager. Her responsibilities include client research and analysis, as well as public relations and business development. Before joining ndp | analytics, Mary was an Account Executive at the Kellen Company where she provided full service management, including government affairs work and strategic consulting, to trade associations in the payments and food-business industries. Mary is currently working towards her Master's in Applied Economics at the University of Maryland and received her Bachelor's in International Relations and French from State University of New York (SUNY) Geneseo.

About Us ndp | analytics is a strategic research firm that specializes in economic analysis of public policy and legal issues. Our services include economic impact studies, business impact analyses, cost-benefit analyses, statistics, and data construction. Our analytical frameworks are data-driven and are supported by economic fundamentals, robust, transparent, and defensible. We excel in supporting organizations for advocacy, government and industry relations, public affairs campaigns, and strategic initiatives. Clients of ndp | analytics include trade associations, coalitions, financial institutions, law firms, U.S. and foreign corporations, and multinational organizations. Our work has been prominently cited in the 2011 Economic Report of the President to the Congress, the media, reports from government agencies, Congressional testimonies, and by Congressional leaders.