Lessons From The Impact Of Price Regulation On The Pricing Of Anticancer Drugs In Germany

Drug spending worldwide has increased dramatically in recent years,¹ with total expenditures projected to reach $1.4 trillion worldwide by the end of 2020.² Such spending has historically been highest in the US, where per capita pharmaceutical spending has almost doubled, from $665 in 2002 to $1,220 in 2016,¹ a phenomenon mainly driven by higher prices rather than by increased prescribing rates.³

Anticancer drugs now account for approximately 12 percent of US pharmaceutical spending,^1,4 with expenditures predicted to grow by 12–15 percent annually during the next five years.^4,5 Inflation- and benefit-adjusted launch prices of anticancer drugs increased by 10 percent annually from 1995 to 2013 in the US,⁶ with an additional average 18 percent markup within eight years after launch,⁷ raising concerns about the sustainability of price growth.

At this time, there are no national-level mechanisms for directly addressing high US drug prices. These prices have been linked to reduced access^8,9 and worse patient adherence^10–12 and, as a consequence, to negative health outcomes¹⁰ and increased health care spending.¹¹ Thus, high brand-name prescription drug prices continue to be the subject of ongoing political¹³ and academic^9,14,15 discussions. In response, pharmaceutical industry representatives have raised concerns that any restraints on drug pricing might either hamper investment in innovation¹⁴ or delay or reduce patients’ access to certain drugs.¹⁶

European countries began implementing regulatory interventions to temper growing pharmaceutical spending as early as the 1990s and have served as models for proposed changes elsewhere.¹⁷ Some countries—most prominently England—conduct economic evaluations that compare incremental cost-effectiveness ratios and willingness-to-pay thresholds to inform binary (yes/no) coverage and reimbursement decisions.¹⁸ Other countries, such as Austria, use external reference pricing and base their prices on those charged in comparable markets.¹⁹ Finally, countries such as Germany apply two-stage approaches in which evidence-based clinical benefit assessments precede price negotiations.¹⁸ The German approach is particularly relevant for potential policy interventions in the US, as Germany also has a multipayer health insurance system and a large domestic pharmaceutical industry.^20,21 Pharmaceutical spending levels in Germany are also among the highest in Europe, at $800 per capita in 2016 versus a European average of $517 per capita, but they are still much lower than in the US.¹

In 2011 Germany launched a major drug pricing reform in response to steadily rising pharmaceutical spending. The 2011 German Pharmaceutical Market Restructuring Act (Arzneimittelmarktneuordnungsgesetz, or AMNOG) was introduced to align prices and reimbursement more closely with expected treatment benefits, with the stated goals of ensuring patients’ access to the best available medicines and providing reliable conditions that promote innovation.²²

AMNOG’s postmarket comparative effectiveness–based price regulation takes place after a product has been authorized for use in the European Union by the European Medicines Agency. Under AMNOG, the manufacturer sets prices freely during a drug’s first year on the market. During this time, additional therapeutic benefits relative to existing standards of care are formally assessed by a nonprofit, nongovernmental research body, the Institute for Quality and Efficiency in Health Care, and by a regulatory agency, the Federal Joint Committee, the highest decision-making body of the German health care system’s joint self-government of professionals, hospitals, and insurers.

Prices are negotiated between manufacturers and the umbrella organization of German statutory health insurers on the basis of this benefit assessment.²³ For drugs without sufficient clinical evidence of therapeutic benefit that surpasses the standard of care (for example, if a clinical or patient-relevant advantage cannot be seen or if the evidence is of inadequate quality), an upper bound for reimbursement is set at the price of the existing standard of care. Negotiated prices of drugs found to have additional therapeutic benefits (categorized as not quantifiable, minor, considerable, or major) may include reimbursement premiums above the price of the existing standard of care. With greater additional therapeutic benefit, negotiated price premiums can be enhanced to a commensurate degree.²⁴ Because additional benefits assigned to new products may vary across patient subgroups (such as patients older or younger than a certain age), negotiated prices may constitute mixed prices that weight different levels of added benefit by the sizes of the respective patient populations. If negotiations fail, a price is set by arbitration.^25,26

Previous research on the AMNOG process has compared decision making between the Federal Joint Committee and other institutions,^27–30 assessed the role of specific types of evidence such as quality of life in AMNOG process outcomes,³¹ and analyzed determinants of price negotiation results.²⁴ We add to this body of knowledge by evaluating whether the AMNOG process has led to drug pricing more aligned with clinical benefit and estimating how much is being paid for an added unit of health gain. Specifically, if the AMNOG process is working as intended, one would expect to see a stronger positive correlation between drug prices and added clinical benefit after AMNOG’s implementation.

To date, there has been no evidence of how AMNOG’s introduction has affected the initial price-setting behavior of manufacturers. A particular concern is that the AMNOG process might incentivize manufacturers to increase launch prices to offset anticipated future discounts from price negotiations.^32,33 This study empirically answers these questions and provides policy guidance for German authorities, as well as for other countries that may consider adopting aspects of Germany’s system for drug pricing.

Study Data And Methods

Sample Selection

We focused on the market for anticancer drugs because of its substantial size and because overall survival and progression-free survival are common primary endpoints used to measure incremental health benefits and conduct comparative effectiveness analyses in oncology.³⁴

We identified 110 anticancer drugs launched in Germany between 2002 and 2017, using annual reports on the German prescription drug market,³⁵ and categorized them as having launched in the period before AMNOG (2002–10) or during the AMNOG period (2011–17). We restricted the sample both to those products that were demonstrated to extend patients’ overall or progression-free survival and to each product’s first European Medicines Agency–approved indication.⁶ We excluded vaccines ($n=2$), palliative treatments ($n=2$), diagnostics ($n=1$), and products primarily intended for the treatment or prevention of adverse effects ($n=5$).

Data Sources And Extraction

We extracted data from several sources, including the European Medicines Agency’s website,³⁶ the Rote Liste (a comprehensive directory of prescription drug prices in Germany),³⁷ the Federal Joint Committee’s website,³⁸ and a cancer registry maintained by the Robert Koch Institute.³⁹ We excluded products with authorizations based on single-group trials ($n=19$), with trials showing no survival benefits ($n=5$), in which survival was not assessed ($n=11$), and in which overall or progression-free survival benefits were reported as probabilities instead of median times ($n=4$). We also excluded products for which we could not find price data ($n=4$). The final analytic data set contained 57 drugs and information about each drug’s incremental health benefits and incremental treatment costs both at launch and after price negotiation, as well as several control variables.

Lists of the drugs included in and excluded from the study sample, a detailed description of the data collection process, an overview of all control variables, and details on how these variables may have influenced a drug’s value are in the online appendix.⁴⁰

Analytic Plan

On the basis of available data, we estimated the amount paid for an additional life-year gained for products that were assessed as having added benefit over the contemporaneous standard of care during the AMNOG period.

Next, we conducted regression analysis to assess the behavior of the manufacturers when they set prices at drug launch. We modeled the relationship between the dependent variable (incremental treatment costs for anticancer drugs in Germany at launch) and two independent variables (life-months gained in either progression-free or overall survival before and after AMNOG went into effect, captured by main effects and an interaction term, respectively, and the year of drug launch). For products launched during the AMNOG period (2011 or later), we also modeled the relationship between incremental treatment costs after price negotiation (as the dependent variable) and life-months gained in progression-free or overall survival and the year of drug launch (as independent variables).

Because each of our dependent variables had a right-skewed distribution, we estimated Poisson regressions with robust standard errors, which are preferred to log-linear regressions in such settings.^41–43 However, Poisson regression requires dependent variables to be greater or equal to 0. We therefore dropped one observation with negative incremental treatment costs after price negotiation (in this unique case, the treatment costs associated with the new product were below those of the comparator therapy). Given our small sample size, we included control variables sequentially and separately, following the method employed by David Howard and colleagues.⁶ All analyses were performed using Stata/SE 15.1. Estimating equations used in the regressions are in the appendix.⁴⁰

To ensure the robustness of our findings, we performed five separate sensitivity analyses, as discussed in the appendix.⁴⁰

Limitations

Our analyses were subject to several limitations, including reliance on aggregated and averaged data to estimate how much payers spent on anticancer products. Because we compared the highest level of benefit reported in European Medicines Agency authorization reports with negotiated prices that weighted different added benefits (when present) by the sizes of constituent patient subpopulations, our results likely underestimate how much German payers spent on a life-year gained.

In addition, the sample size was modest (totaling fifty-seven drugs), preventing us from including all potentially relevant control variables simultaneously in regression analyses and limiting our ability to detect associations between variables. As such, we could include the year of launch as only a linear time trend in our regression models. Future analyses of larger samples might include launch years as independent dummy variables to obtain greater flexibility in accounting for differences over time.

We also were unable to account for confidential supplemental rebates that manufacturers and individual statutory health insurers may have negotiated, and we did not account for costs incidental to the use of therapies, such as inpatient or outpatient care and treatment of adverse effects. Although data on these costs are available for products assessed through the AMNOG process, the same information is not available for products launched before 2011, nor is it the case that comparators used in the European Medicines Agency’s evaluations were the same as those used in the AMNOG process.

In addition, we relied on data on progression-free survival, a surrogate measure, in cases in which data on overall survival were not available. Although the correlation between progression-free survival and overall survival has been found to be low or modest in many settings,⁴⁴ progression-free survival has been considered by some as a valid clinical endpoint in certain clinical settings, such as treatment for ovarian and metastatic colon cancer.^6,45

Finally, our sample was not representative of drugs for rare cancers, which are treated differently than other products in the AMNOG process.

Study Results

Our final sample included fifty-seven anticancer drugs, fourteen of which launched in the period before AMNOG and forty-three of which launched in the AMNOG period. The average annual incremental treatment costs at launch were $51,127 over the entire sample (exhibit 1). These costs were smaller ($p=0.02$; data not shown) in the period before AMNOG (mean, $29,417; data not shown) than in the AMNOG period (mean, $58,195; data not shown), but were not different ($p=0.20$; data not shown) from annual incremental treatment costs based on negotiated prices in the AMNOG period ($43,953; exhibit 1).

The mean additional overall or progression-free survival gain of anticancer drugs launched in Germany between 2002 and 2017 was 6.4 life-months (exhibit 1), with a minimum of 0.2 life-months and a maximum of 49.1 life-months (data not shown; median, 3.5 life-months; interquartile range, 2.4–5.8 life-months). The average baseline overall or progression-free survival for the sample was 10.2 life-months (exhibit 1; median, 9.1 life-months; interquartile range, 5.0–14.5 life-months; data not shown). Gains were stable during the observation period (exhibit 1).

Comparing the incremental treatment costs of newly launched anticancer drugs both at launch and after price negotiation with their associated overall or progression-free survival life-months gained revealed that during the AMNOG period, incremental treatment costs were positively associated with overall or progression-free survival life-months gained. Incremental treatment costs after negotiation were consistently lower than incremental treatment costs at launch after AMNOG went into effect (exhibit 2). The average difference between annual incremental treatment costs at launch versus after negotiation was $14,242, representing a relative decrease in incremental treatment costs over the period of benefit assessment and negotiation of 24.5 percent (data not shown). In the period before AMNOG, by contrast, there was no evidence of an association between incremental treatment costs and overall or progression-free survival life-months gained.

SOURCE Authors’ analysis of data from various sources (details are in the appendix; see note 40 in text). NOTE Data include all fifty-seven drugs in the study sample, in 2018 prices.

Estimated average incremental treatment costs per overall or progression-free survival life-month gained at launch were $16,041 in the period before AMNOG and $15,263 in the AMNOG period, which decreased to $11,476 after price negotiation (data not shown). Neither the average incremental treatment cost per overall or progression-free survival life-month gained at launch or after negotiation in the AMNOG period was different from the average incremental treatment cost per overall or progression-free survival life-month gained at launch in the period before AMNOG ($p=0.92$ and $p=0.50$, respectively).

Because the estimates of cost per overall or progression-free survival life-month gained were strongly influenced by a single outlier, the acute lymphoblastic leukemia treatment inotuzumab (Besponsa), we excluded it in subsequent estimation exercises. Differentiating by added benefit status in the AMNOG period revealed average additional treatment costs at launch of $12,267 per overall or progression-free life-month gained for products without an added benefit ($n=8$), which decreased to $6,994 after price negotiation (data not shown). Given that negotiated prices for products that did not yield added benefit resulted in treatment costs equivalent to those of the current standard of care,²⁴ we estimated that pharmaceutical manufacturers in Germany received an average of $6,994 per overall or progression-free life-month gained for already-established anticancer drugs during this period.

For products assessed to have added benefit ($n=34$), median additional treatment costs were $11,191 per overall or progression-free survival life-month gained at launch and decreased to $8,690 after price negotiation (data not shown). An additional overall or progression-free survival life-month gained for products with added benefit was therefore reimbursed at an average of $1,696 ($20,352 per overall or progression-free survival life-year gained).

These estimates are based on manufacturer prices, net of mandatory rebates. However, actual spending of the statutory health insurer is at least 19 percent higher in practice as a result of value-added tax and wholesaler and pharmacy margins. This implies an extrapolated cost of $24,219 per additional life-year gained in overall or progression-free survival for products found to have added benefit.

Importantly, these estimates are based on the assumptions that the comparator costs for products with and without added benefit were comparable and that the European Medicines Agency and Federal Joint Committee had the same assessment for products with an added benefit. Both of these assumptions were confirmed via t-tests with data collected from the Federal Joint Committee’s website ($p=0.23$ and $p=0.26$, respectively; see the appendix for additional detail).⁴⁰

The regression estimates presented in exhibit 3 suggest that in the AMNOG period, each additional overall or progression-free survival life-month gained was associated with an average increase in incremental treatment costs of 3.0 percent after price negotiation. By contrast, at drug launch, we did not find evidence of an association between incremental treatment costs and the number of overall or progression-free survival life-months gained, neither during the full period of observation nor in the AMNOG period alone.

In addition, we found that although treatment costs at launch increased by an average of 10.6 percent per year during the full period of observation, there was no evidence of such a time trend in negotiated prices in the AMNOG period (exhibit 3).

The results of both models in exhibit 3 were largely robust to the inclusion of control variables. Reestimating regression models without outliers, with overall survival data only, without products launched in 2011, or by using log-linear ordinary least squares specifications yielded highly similar results but rendered the coefficient on year insignificant in most models in which incremental treatment cost at launch was the dependent variable. In a sensitivity analysis that included a variable indicating that multiple patient subgroups had been defined and assessed differently, treatment costs after price negotiation decreased by 65.3 percent. This negative coefficient can likely be explained by the common practice of negotiating mixed prices for such products (see the appendix).⁴⁰

Discussion

In response to increases in drug prices during the past few decades, in 2011 Germany passed the Pharmaceutical Market Restructuring Act, known as AMNOG. We found that after AMNOG’s passage, drug prices were more closely aligned with clinical benefits. During the AMNOG period, there was a significant association between overall or progression-free survival and incremental treatment costs after price negotiation, suggesting that the process was successful in linking prices to clinical benefit in a way not previously observed. We did not find evidence consistent with manufacturers increasing launch prices to offset anticipated future discounts predicated to result from negotiation.

Among the drugs in our sample, price negotiations decreased incremental treatment costs by an average of 24.5 percent (data not shown). In addition, price negotiations offset a positive time trend in launch prices,⁶ although sensitivity analyses suggest that this time trend was driven by a small number of products with exceptionally high prices.

Many of the goals associated with the implementation of AMNOG were met for anticancer drugs.

Many of the goals associated with the implementation of AMNOG were met for anticancer drugs. Importantly, the prices of these drugs became increasingly aligned with treatment benefit. These results are particularly reassuring in the context of prior research showing that AMNOG did not lead manufacturers to withdraw clinically important medications from the German market,²⁵ suggesting that the change in pricing did not affect patients’ access in clinically meaningful ways. Further research should more closely examine the extent to which the implementation of AMNOG has affected uptake of pharmaceutical products, prescribing rates, and patient outcomes—factors likely to be of key interest to health care policy makers in Germany and other countries.

The estimate of a median cost of $1,696 per additional overall or progression-free survival life-month gained for anticancer drugs with added benefit includes some products with mixed prices and is not adjusted for quality of life. As such, a direct comparison to other countries’ willingness-to-pay thresholds, usually measured in dollars per quality-adjusted life-year (a life-year gained in full health) is difficult. Nonetheless, as a benchmark, we note that England’s willingness-to-pay threshold for one quality-adjusted life-year in end-of-life treatments (primarily cancer therapies) is approximately $63,595 (£50,000),⁴⁶ which is higher than our (most likely underestimated) estimate of $24,219 (extrapolated out to a year and accounting for value-added tax).

Summary statistics indicate that treatment costs associated with new drugs kept increasing beyond the introduction of AMNOG. However, comparing the growth in overall health care spending with the growth in pharmaceutical-specific spending before and after the introduction of AMNOG presents a different picture. In the seven years before AMNOG (2004–10), standardized health care spending grew by 23.6 percent, while spending growth for pharmaceuticals was much more rapid, increasing by 32.3 percent during the same period. In the seven years since the implementation of AMNOG (2011–17), standardized overall health care spending and pharmaceutical spending grew at similar rates (26.7 percent and 27.4 percent, respectively).⁴⁷

Although the introduction of prices more closely aligned with treatment benefits suggests progress toward AMNOG’s goals, it might not necessarily be associated with a decrease in spending. As Karl Claxton has argued, value-based pricing “would lead to higher prices for some products and involve a reallocation of revenue from products which are less valuable to those products which are more valuable.”^48(p547) Prices more closely aligned with treatment benefits should incentivize manufacturers to produce innovative pharmaceuticals with high value and might, in the long run, increase overall pharmaceutical spending levels and growth, albeit for more clinically important products with better patient outcomes.⁴⁸

Policy Implications

Our findings are relevant for other countries facing increasing prescription drug prices. In the US, cancer drug prices are set by pharmaceutical manufacturers and are not systematically negotiated by the government in connection with a product’s clinical benefit. At this time, US policy makers are discussing a number of ways to reform US drug pricing, with different approaches being proposed to change the way Medicare pays for medications through its Part B and Part D programs. Individually, states are also considering new options for determining how to pay for expensive but important new medications through Medicaid.⁴⁹ Germany’s AMNOG experience suggests a model for what might happen if policy makers are successful in introducing a drug price negotiation system in which prices are more closely linked to clinical benefit. This study also shows that negotiation can lead to further reductions in cancer drug prices and that these reductions appear to have offset time trends in drug price increases not correlated with clinical benefit. Another feature of the AMNOG process is that price negotiation occurs after a drug has already entered the market, and thus should not delay patient access to medicines.

In the absence of further policy changes, application of an AMNOG-like system of evidence-based price negotiations in the US would likely not lead to the same changes in oncology drug prices as those observed in Germany. For example, cancer drugs remain a protected drug class for the Medicare Part D outpatient drug insurance program, which covers patients older than age sixty-five. Because Part D programs must include all drugs in protected classes in their formularies, manufacturers may feel less pressure to bring their prices in line with clinical benefit. Many states also have rules limiting the flexibility of private insurers to exclude cancer drugs from coverage under certain conditions.⁵⁰ In addition, the German statutory health insurers jointly negotiate prices on behalf of all payers, representing a level of coordination that has no analogue in the US, where negotiations between manufacturers and payers are fragmented, secretive, and often managed by pharmacy benefit manager intermediaries.⁵¹ Therefore, policy makers in the US would need to revisit coverage rules as well as rules governing how prices are negotiated to effectively implement similar policies.

Conclusion

AMNOG’s introduction of benefit assessments followed by price negotiations for newly authorized prescription drugs in Germany led to anticancer drug prices that were more closely aligned with treatment benefit. Other countries, including the US, should consider components of Germany’s AMNOG system as they consider ways to address rising pharmaceutical spending.

ACKNOWLEDGMENTS

Victoria D. Lauenroth has been employed with AstraZeneca Germany since October 2019, after the initial manuscript was submitted but before revisions were finalized. Aaron S. Kesselheim and Ameet Sarpatwari are supported by Arnold Ventures and the Harvard-MIT Center for Regulatory Science. Ariel D. Stern is supported by the Kauffman Junior Faculty Fellowship. The authors are grateful to Mats Terwiesch, Melissa Ouellet, and Lila Kelso for excellent research assistance.

NOTES

• 1 Organization for Economic Cooperation and Development. OECD data: pharmaceutical spending [Internet]. Paris: OECD; 2019 [cited 2020 Apr 27]. Available from: https://data.oecd.org/healthres/pharmaceutical-spending.htm Google Scholar
• 2 IMS Institute for Healthcare Informatics. Global medicines use in 2020 [Internet]. Danbury (CT): IMS Health; 2015 Nov [cited 2020 Apr 27]. Available from: https://www.iqvia.com/-/media/iqvia/pdfs/institute-reports/global-medicines-use-in-2020 Google Scholar
• 3 Sarnak D, Squires D, Kuzmak G, Bishop S. Paying for prescription drugs around the world: why is the U.S. an outlier? Issue Brief (Commonw Fund). 2017;2017:1–14. Medline, Google Scholar
• 4 IQVIA. Global oncology trends 2018 [Internet]. Durham (NC): IQVIA; 2018 May 24 [cited 2020 Apr 27]. Available from: https://www.iqvia.com/institute/reports/global-oncology-trends-2018 Google Scholar
• 5 Chandra A, Garthwaite C, Stern AD. Characterizing the drug development pipeline for precision medicines [Internet]. Cambridge (MA): National Bureau of Economic Research; 2017 Nov [last updated 2018 May; cited 2020 Apr 27]. (NBER Working Paper No. 24026). Available from: http://www.nber.org/papers/w24026.pdf Google Scholar
• 6 Howard DH, Bach PB, Berndt ER, Conti RM. Pricing in the market for anticancer drugs. J Econ Perspect. 2015;29(1):139–62. Crossref, Medline, Google Scholar
• 7 Gordon N, Stemmer SM, Greenberg D, Goldstein DA. Trajectories of injectable cancer drug costs after launch in the United States. J Clin Oncol. 2018;36(4):319–25. Crossref, Medline, Google Scholar
• 8 Experts in Chronic Myeloid Leukemia. The price of drugs for chronic myeloid leukemia (CML) is a reflection of the unsustainable prices of cancer drugs: from the perspective of a large group of CML experts. Blood. 2013;121(22):4439–42. Crossref, Medline, Google Scholar
• 9 Alexander GC, Ballreich J, Socal MP, Karmarkar T, Trujillo A, Greene Jet al. Reducing branded prescription drug prices: a review of policy options. Pharmacotherapy. 2017;37(11):1469–78. Crossref, Medline, Google Scholar
• 10 Gagne JJ, Choudhry NK, Kesselheim AS, Polinski JM, Hutchins D, Matlin OSet al. Comparative effectiveness of generic and brand-name statins on patient outcomes: a cohort study. Ann Intern Med. 2014;161(6):400–7. Crossref, Medline, Google Scholar
• 11 IMS Health. Avoidable costs in U.S. healthcare [Internet]. Danbury (CT): IMS Health; 2013 Jun [cited 2020 May 26]. Available from: http://offers.premierinc.com/rs/381-NBB-525/images/Avoidable_Costs_in%20_US_Healthcare-IHII_AvoidableCosts_2013%5B1%5D.pdf Google Scholar
• 12 Kirzinger A, Lopes L, Wu B, Brodie M. KFF Health Tracking Poll—February 2019: prescription drugs [Internet]. San Francisco (CA): Henry J. Kaiser Family Foundation; 2019 Mar 1 [cited 2020 Apr 27]. Available from: https://www.kff.org/health-costs/poll-finding/kff-health-tracking-poll-february-2019-prescription-drugs/ Google Scholar
• 13 White House. President Donald J. Trump’s blueprint to lower drug prices [Internet]. Washington (DC): White House; 2018 May 11 [cited 2020 Apr 27]. Available from: https://www.whitehouse.gov/briefings-statements/president-donald-j-trumps-blueprint-lower-drug-prices/ Google Scholar
• 14 Kesselheim AS, Avorn J, Sarpatwari A. The high cost of prescription drugs in the United States: origins and prospects for reform. JAMA. 2016;316(8):858–71. Crossref, Medline, Google Scholar
• 15 Prasad V, De Jesús K, Mailankody S. The high price of anticancer drugs: origins, implications, barriers, solutions. Nat Rev Clin Oncol. 2017;14(6):381–90. Crossref, Medline, Google Scholar
• 16 Mason A, Drummond M, Ramsey S, Campbell J, Raisch D. Comparison of anticancer drug coverage decisions in the United States and United Kingdom: does the evidence support the rhetoric? J Clin Oncol. 2010;28(20):3234–8. Crossref, Medline, Google Scholar
• 17 Drummond M. Twenty years of using economic evaluations for drug reimbursement decisions: what has been achieved? J Health Polit Policy Law. 2013;38(6):1081–102. Crossref, Medline, Google Scholar
• 18 Paris V, Belloni A. Value in pharmaceutical pricing [Internet]. Paris: Organization for Economic Cooperation and Development; 2013 Jul 11 [cited 2020 Apr 27]. Available from: http://www.oecd-ilibrary.org/social-issues-migration-health/value-in-pharmaceutical-pricing_5k43jc9v6knx-en Google Scholar
• 19 Rémuzat C, Urbinati D, Mzoughi O, El Hammi E, Belgaied W, Toumi M. Overview of external reference pricing systems in Europe. J Mark Access Health Policy. 2015;3(1):27675. Crossref, Google Scholar
• 20 Lauterbach K, McDonough JE, Seeley E. Germany’s model for drug price regulation could work in the US. Health Affairs Blog [blog on the Internet]. 2016 Dec 29 [cited 2020 Apr 27]. Available from: https://www.healthaffairs.org/do/10.1377/hblog20161229.058150/full/ Google Scholar
• 21 Robinson JC, Ex P, Panteli D. Single-payer drug pricing in a multipayer health system: does Germany offer a model for the US? Health Affairs Blog [blog on the Internet]. 2019 Mar 22 [cited 2020 Apr 27]. Available from: https://www.healthaffairs.org/do/10.1377/hblog20190318.475434/full/ Google Scholar
• 22 German Bundestag. Draft legislation: Pharmaceutical Market Restructuring Act [Internet]. Bonn: German Bundestag; 2010 Jul 6 [cited 2020 May 26]. German. Available from: http://dip21.bundestag.de/dip21/btd/17/024/1702413.pdf Google Scholar
• 23 Federal Ministry of Justice and Consumer Protection. Agreements between the National Association of Health Insurance Funds and pharmaceutical manufacturers on reimbursement amounts for drugs [Internet]. Bonn: The Ministry; [cited 2020 May 26]. German. Available from: https://www.gesetze-im-internet.de/sgb_5/__130b.html Google Scholar
• 24 Lauenroth VD, Stargardt T. Pharmaceutical pricing in Germany: how is value determined within the scope of AMNOG? Value Health. 2017;20(7):927–35. Crossref, Medline, Google Scholar
• 25 Stern A, Pietrulla F, Herr A, Kesselheim AS, Sarpatwari A. The impact of price regulation on the availability of new drugs in Germany. Health Aff (Millwood). 2019;38(7):1182–7. Go to the article, Google Scholar
• 26 Ludwig S, Dintsios CM. Arbitration board setting reimbursement amounts for pharmaceutical innovations in Germany when price negations between payers and manufacturers fail: an empirical analysis of 5 years’ experience. Value Health. 2016;19(8):1016–25. Crossref, Medline, Google Scholar
• 27 Fischer KE, Stargardt T. Early benefit assessment of pharmaceuticals in Germany: manufacturers’ expectations versus the Federal Joint Committee’s decisions. Med Decis Making. 2014;34(8):1030–47. Crossref, Medline, Google Scholar
• 28 Fischer KE, Heisser T, Stargardt T. Health benefit assessment of pharmaceuticals: an international comparison of decisions from Germany, England, Scotland, and Australia. Health Policy. 2016;120(10):1115–22. Crossref, Medline, Google Scholar
• 29 Peinemann F, Labeit A. Varying results of early benefit assessment of newly approved pharmaceutical drugs in Germany from 2011 to 2017: a study based on Federal Joint Committee data. J Evid Based Med. 2019;12(1):9–15. Crossref, Medline, Google Scholar
• 30 Griffiths EA, Macaulay R, Vadlamudi NK, Uddin J, Samuels ER. The role of noncomparative evidence in health technology assessment decisions. Value Health. 2017;20(10):1245–51. Crossref, Medline, Google Scholar
• 31 Blome C, Augustin M, Metin H, Lohrberg D. Four years of early benefit assessment of new drugs in Germany: a qualitative study on methodological requirements for quality of life data. Eur J Health Econ. 2017;18(2):181–93. Crossref, Medline, Google Scholar
• 32 Congressional Budget Office. How the Medicaid rebate on prescription drugs affects pricing in the pharmaceutical industry [Internet]. Washington (DC): CBO; 1996 Jan 1 [cited 2020 Apr 27]. (CBO Papers). Available from: https://www.cbo.gov/publication/14944 Google Scholar
• 33 Conti RM, Bach PB. Cost consequences of the 340B Drug Discount Program. JAMA. 2013;309(19):1995–6. Crossref, Medline, Google Scholar
• 34 European Medicines Agency. Guideline on the evaluation of anticancer medicinal products in man [Internet]. Amsterdam: EMA; 2017 Nov 20 [last updated 2018 Apr 1; cited 2020 Apr 27]. Available from: https://www.ema.europa.eu/en/evaluation-anticancer-medicinal-products-man Google Scholar
• 35 Schwabe U, Paffrath D, Ludwig W-D, Klauber J. Arzneiverordnungs-Report 2018. Berlin: Springer; 2018. German. Crossref, Google Scholar
• 36 European Medicines Agency. Medicines [Internet]. Amsterdam: EMA; 2019 [cited 2020 Apr 27]. Available from: https://www.ema.europa.eu/en/medicines Google Scholar
• 37 Rote Liste 2018. Arzneimittelverzeichnis für Deutschland. Frankfurt am Main: Rote Liste Service GmbH; 2018. German. Google Scholar
• 38 Gemeinsamer Bundesausschuss. Verfahren der Nutzenbewertung nach § 35a SGB V [Internet]. Berlin: G-BA; 2019 [cited 2020 Apr 27]. German. Available from: https://www.g-ba.de/bewertungsverfahren/nutzenbewertung/ Google Scholar
• 39 Zentrum für Krebsregisterdaten. Datenbankabfrage [Internet]. Berlin: Robert Koch Institute; 2019 [cited 2020 Apr 27]. German. Available from: https://www.krebsdaten.de/Krebs/DE/Datenbankabfrage/datenbankabfrage_stufe1_node.html Google Scholar
• 40 To access the appendix, click on the Details tab of the article online.
• 41 Gould W. Use Poisson rather than regress; tell a friend. The Stata Blog [blog on the Internet]. 2011 Aug 22 [cited 2020 Apr 27]. Available from: https://blog.stata.com/2011/08/22/use-poisson-rather-than-regress-tell-a-friend/ Google Scholar
• 42 Wooldridge JM. Econometric analysis of cross section and panel data. 2nd ed. Cambridge (MA): MIT Press; 2010. Google Scholar
• 43 Cameron AC, Trivedi PK. Microeconometrics using Stata. College Station (TX): Stata Press; 2010. Google Scholar
• 44 Haslam A, Hey SP, Gill J, Prasad V. A systematic review of trial-level meta-analyses measuring the strength of association between surrogate end-points and overall survival in oncology. Eur J Cancer. 2019;106:196–211. Crossref, Medline, Google Scholar
• 45 Bast RC, Thigpen JT, Arbuck SG, Basen-Engquist K, Burke LB, Freedman Ret al. Clinical trial endpoints in ovarian cancer: report of an FDA/ASCO/AACR Public Workshop. Gynecol Oncol. 2007;107(2):173–6. Crossref, Medline, Google Scholar
• 46 National Institute for Health and Care Excellence, Centre for Health Technology Evaluation. Consultation Paper: Value based assessment of health technologies [Internet]. London: NICE; [cited 2020 May 26]. Available from: https://www.nice.org.uk/Media/Default/About/what-we-do/NICE-guidance/NICE-technology-appraisals/VBA-TA-Methods-Guide-for-Consultation.pdf Google Scholar
• 47 Federal Health Reporting. Current health expenditure in millions of euro (year, provider, function, financing agent) [Internet]. Berlin: Robert Koch Institut; 2019 [cited 2020 Jun 15]. German. Available for download from: http://www.gbe-bund.de Google Scholar
• 48 Claxton K. OFT, VBP: QED? Health Econ. 2007;16(6):545–58. Crossref, Medline, Google Scholar
• 49 Trusheim MR, Cassidy WM, Bach PB. Alternative state-level financing for hepatitis C treatment—the “Netflix Model.” JAMA. 2018;320(19):1977–8. Crossref, Medline, Google Scholar
• 50 Ramsey SD. How state and federal policies as well as advances in genome science contribute to the high cost of cancer drugs. Health Aff (Millwood). 2015;34(4):571–5. Go to the article, Google Scholar
• 51 Carrier M. A six-step solution to the PBM problem. Health Affairs Blog [blog on the Internet]. 2018 Aug 30 [cited 2020 Apr 27]. Available from: https://www.healthaffairs.org/do/10.1377/hblog20180823.383881/full/ Google Scholar