The seminal work of Wennberg and Gittelsohn in 1973 emphasised the importance of health information for informed decision-making. This led to the creation of the Dartmouth Health Atlas in

1996, which has become an important resource for monitoring health services in the USA. The Dartmouth Health Atlas research revealed the existence of variation in health care without benefit

to patients, and the dependence of health care use on local resource supply. Similar initiatives emerged around the world, from the UK to Asia. The availability of administrative data has

become essential for evaluating health service delivery and for informing health economic analysis and policy decisions. Access to data depends on the organisation of the health system, with

billing patterns. This variability can be attributed to the lack of clear guidelines and the complexity of billing codes. Incentives to charge incorrect rates influence billing, adding a

further variance component to the variance in care that cannot be subtracted from the total variance at the level of a health insurance fund and distorting the results. In certain

environments the quality of data on care is so variable that a sound conclusions for health policy decisions represent a great challenge. THE GOLDEN AGE OF DATA-DRIVEN DECISION MAKING In

1973, in their seminal publication in Science, Wennberg and Gittelsohn highlighted the importance of health information about total populations for sound decision-making and planning [1].

Their work eventually led to the Dartmouth Health Atlas in 1996, which has become a pivotal source to monitor and manage health service delivery in the United States of America [2]. The

research led by the Dartmouth Health Atlas team has several key findings: Variation in healthcare delivery exists and has no benefit for patients. If the evidence supporting an intervention

is weak, variation in care reflects physicians’ rather than patients’ preferences. The extent of health service utilization depends on the local supply of resources, and more health care

might not be better [3]. These findings from the USA have had a strong impact and led to similar initiatives in the United Kingdom, Europe, South America, Asia, and Oceania in the last two

decades [3, 4]. DATA AVAILABILITY AS A CATALYST FOR RESEARCH An important driver for the development of the Dartmouth Health Atlas was the availability of Medicare and Medicaid data,

providing a comprehensive coverage of health service delivery data for the elderly and the low-income population in the United States [3, 5]. To date, administrative data have become an

indispensable resource to assess service delivery in healthcare. In the literature the terms “health care utilization data”, “billing records”, “administrative claims data”, or simply

“claims data” are often used interchangeably to describe all sorts of administrative data and billing purposes derived from the health care sector [6]. Based on Wennberg’s ground-breaking

work on analysing the variability of care and its impact on healthcare, the use of administrative data to measure the quality of care has taken off. Hardly any health economic analysis can

do without this data (summarised in [7]). Political decisions on the organisation of the healthcare system are based on this data. Recently, there have been calls for the assessment of the

effectiveness of drugs to be based not only on the results of controlled studies, but also on the effects observed in routine clinical practice [8]. DATA ACCESS AND POTENTIAL Access to

administrative data depends heavily on the organisation of the healthcare system (summarised in [9]). The more centralised the control system is, the more likely it is that care data can be

collected uniformly and comprehensively. In addition to the conditions in the USA, where the introduction of Medicare and Medicaid in 1965 and the availability of Surveillance, Epidemiology,

and End Results (SEER) [5] created important circumstances for the work of Wennberg and colleagues and many other initiatives including cancer surveillance [10], the Danish healthcare

system and the British healthcare system following the introduction of the National Health System NHS in 1948 should be mentioned in Europe [11]. In Denmark, the establishment of an agency

that oversaw the agreement between industry partners and service providers on the use of a uniform software standard in 1994 led to the development of the Danish Quality Programme

(summarised in [12]). From a methodological point of view, the use of administrative care data is associated with challenges. The harmonisation and standardisation of data collection and

formats is an important factor for the validity of the statements [9]. It is therefore no coincidence that healthcare systems that have adopted a top-down approach to planning and monitoring

healthcare have been able to implement quality programmes more easily. The collection of real-world data is also playing an increasingly important role in assessing the efficacy and safety

of new therapies. It has long been recognised that the results of controlled clinical trials, which play a crucial role in the approval of a new medicine, can only partially reflect the

benefits in routine clinical practice. The collection of real-world data has become increasingly important, not only for post-marketing surveillance and pharmacovigilance, but also for

efficient trial design, drug labelling, and the approval of new therapies. Recently, even regulatory authorities have begun to require it prior to approval [13]. ORGANISATION OF HEALTHCARE

OR FORM FOLLOWS FUNCTION The Swiss healthcare system enjoys an excellent reputation by international standards [14]. In contrast to the USA or the UK, the system has a federal structure. The

planning and management of the healthcare system is highly decentralised and regulated at cantonal level in important respects [15]. As a result, 26 different healthcare systems with

different regulations provide healthcare in Switzerland, a relatively small country with a population of around 9 million. The federal structure has the great advantage that the local needs

of the population can be better considered. However, the implementation of national healthcare programmes poses a challenge [16]. In many cases, it is the different regulations in the

cantons or the different processes that hinder nationwide health monitoring. This also applies to national surveys of the quality of care. In Switzerland and other countries, attempts to

implement the Wennberg and Dartmouth Health Atlas approaches have been made. In Switzerland, such an initiative has been recently re-launched, but still faces several problems [17]. In

contrast to the prevailing care structures, Swiss health insurance funds insure patients across all cantons [15]. This makes them an important player in the healthcare system, with not only

a cantonal but also a national view of care. The large health insurance funds in particular have become increasingly involved in the health policy debate and established quality programmes

in the wake of the cost trend in the Swiss healthcare system. Billing data is used in co-operation with the health insurance funds for many questions of health services research

[18,19,20,21,22]. Research with this data has the advantage that statements can be made across the cantons. However, as the health insurance funds do not have access to the clinical data of

their policyholders, the possibilities for analysing the causes of cost variability between service providers and different care structures are limited [19, 20]. Research-based joint

ventures between health insurance funds and service providers, in which the cost data is linked to the clinical data of the insured persons of this service provider, are currently the only

way to carry out further investigations into cost variability. However, these analyses are complex and must be well justified for data protection reasons [21, 22]. In the United Kingdom, the

introduction of a retinal thickness threshold of >400 microns for the treatment of diabetic macular oedema with anti-VEGF drugs, based on a medico-economic analysis within a NICE

guideline [23], has led to a vigorous debate about the optimal setting of treatment standards based on controlled clinical trials [24,25,26]. The analysis of real-world data demonstrated

that this treatment threshold and delayed treatment with anti-VEGF drugs resulted in suboptimal visual outcomes for patients [25, 26]. In the discussion about the remuneration of medical

services in Switzerland, health insurance billing data is often used. They are used to set benchmarks and to better understand and discuss the variability in care. Based on the work of

Wennberg and colleagues, the aim is to identify and analyse conspicuous billing behaviour in the discussion with service providers. When analysing the use and billing of optical coherence

tomography (OCT) in patients with retinal diseases in Switzerland, an interesting situation arose pointing to an additional, presumed driver for variability in the claim data. WHERE’S THE

BEEF?—THE EXAMPLE OF OPTICAL COHERENCE TOMOGRAPHY In ophthalmology, the OCT is a diagnostic procedure typically used to manage patients with cases of neovascular age-related macular

degeneration (nAMD) and diabetic macular oedema (DMO) [27, 28]. All relevant guidelines suggest basing treatment decisions on the results of the OCT scan. While the use of OCT examinations

for therapy monitoring is undisputed, the evidence for binocular examinations when only one eye is initially affected by the retinal disease is unclear. There are no binding recommendations

from guidelines or professional associations [27, 28]. The billing of OCT services is handled differently internationally. OCT examinations are often billed as a lump sum. This means that it

is not clear from the billing data whether one or both eyes were examined. In Switzerland, the OCT examination is billed separately for each eye, which makes it possible to analyse how it

is handled. The analysis of the billing data from the health insurance companies showed a high degree of variability between the individual service providers and billing patterns that were

difficult to interpret. This prompted us to conduct an anonymous survey of the largest providers of retinal disease services in Switzerland to find out how they arrange OCT examinations for

monocular nAMD or DME and how these services are billed. The responses from the 15 largest institutions, which together provide around two thirds of care, revealed interesting behavioural

patterns. VARIABILITY IN CARE We found that four out of five centres performed bilateral OCT at least every three months for both nAMD and DME patients. Half of the centres reported

performing bilateral OCT scans at every visit for DME and two-thirds for nAMD. One-fifth of centres reported performing bilateral OCT scans only when there were clinical signs or symptoms,

such as deterioration in vision or fundoscopic findings suggesting the onset of disease in the untreated eye. BILLING PATTERNS Surprisingly, billing patterns did not match clinical practice

and varied widely. Only one centre billed each bilateral OCT with the correct code for a bilateral examination. Four centres reported that they examined the untreated eye free of charge, and

other centres billed for OCT in the untreated eye only periodically (once or twice a year). In addition, one third of the centres for nAMD and almost half of the centres for DME treated OCT

in the untreated eye as a contingency and only billed for it if the OCT showed progression of the disease. Finally, three centres never billed for OCT for nAMD and one centre never billed

for DME. Instead, they used the cheaper billing code for retinal photography. Billing patterns for treated eyes were consistent across centres. However, a quarter of centres did not always

use the correct code for OCT in the anti-VEGF treated eye, sometimes preferring to use the incorrect, but significantly cheaper, billing code for retinal photography. IS GOOD NOT GOOD

ENOUGH? Half a century of health systems research has contributed significantly to our understanding of health systems and health care [29]. For research with administrative data, such as

health data, to work well, certain framework conditions must be met. Switzerland is an interesting example. As already known from other studies, this survey also shows variability in

healthcare practice [30, 31]. For our case study, we can only speculate about the causes of this variability. The international literature on the correct use of OCT examinations in the

second eye is controversial. In addition, there is a lack of binding guidelines that could serve as a guide for care providers [27, 28]. If guidelines only provide vague recommendations,

this is likely to lead to increased variability. Specialist societies can also make an important contribution to improving data quality. For example, the Swiss Vitreoretinal Group has not

yet commented on how the approximately 300 ophthalmologists in Switzerland who offer intravitreal injections should proceed in this situation. The tariff system on which the health claim

data is based must be unambiguous and provide clear guidelines for billing [32]. Service providers must undertake to adhere to these guidelines. If, for example, disincentives to charge the

wrong rates influence billing, a further variance component is added to the variance of care provision, which cannot be subtracted from the total variance at the level of a health insurance

fund and distorts the results [32]. For a well-founded analysis of the variability of care, it is also important to have information on the initial clinical situation to examine the

heterogeneity in the data [7]. The 21st century is also the digital century in medicine. We will have more and more opportunities to make data-based statements about the quality of

healthcare easier and faster. This vision is often still associated with fears, particularly among service providers. Clear rules and objectives for care analysis are important prerequisites

for reducing these fears. The continuous analysis of the quality of care is an important building block for ensuring the long-term health of the population and providing patients with

optimal care. REFERENCES * Wennberg J, Gittelsohn. Small area variations in health care delivery. Science. 1973;182:1102–8. Article  CAS  PubMed  Google Scholar  * Wennberg JE, Cooper MM.

The Dartmouth Atlas of Health Care in the United States: The Center for the Evaluative Clinical Sciences. Chicago (IL): American Hospital Publishing, Inc. 1996. * Bronner KK, Goodman DC. The

Dartmouth Atlas of Health Care—bringing health care analyses to health systems, policymakers, and the public. Res Health Serv Reg. 2022;1:6. Article  PubMed  PubMed Central  Google Scholar

  * Moorthie S, Peacey V, Evans S, Phillips V, Roman-Urrestarazu A, Brayne C, et al. A scoping review of approaches to improving quality of data relating to health inequalities. Int J

Environ Res Public Health. 2022;19:15874. Article  PubMed  PubMed Central  Google Scholar  * Warren JL, Klabunde CN, Schrag D, Bach PB, Riley GF. Overview of the SEER-Medicare data: content,

research applications, and generalizability to the United States elderly population. Med Care. 2002;40:IV-3-18. Article  PubMed  Google Scholar  * Cadarette SM, Wong L. An introduction to

health care administrative data. Can J Hosp Pharm. 2015;68:232–7. PubMed  PubMed Central  Google Scholar  * Shih YT, Liu L. Use of claims data for cost and cost-effectiveness research. Semin

Radiat Oncol. 2019;29:348–53. Article  PubMed  PubMed Central  Google Scholar  * Eichler HG, Pignatti F, Schwarzer-Daum B, Hidalgo-Simon A, Eichler I, Arlett P, et al. Randomized controlled

trials versus real world evidence: neither magic nor myth. Clin Pharm Ther. 2021;109:1212–8. Article  Google Scholar  * Antonacci G, Whitney J, Harris M, Reed JE. How do healthcare

providers use national audit data for improvement? BMC Health Serv Res. 2023;23:393. Article  PubMed  PubMed Central  Google Scholar  * Mariotto AB, Yabroff KR, Shao Y, Feuer EJ, Brown ML.

Projections of the cost of cancer care in the United States: 2010-20. J Natl Cancer Inst. 2011;103:117–28. Article  PubMed  PubMed Central  Google Scholar  * Disparities of HIa. Atlas of

Variation - 2023 Vision updates. available from https://fingertips.phe.org.uk/vision 2023; last time accessed: 24/04/2024. * Carstensen K, Kjeldsen AM, Lou S, Nielsen CP. The Danish health

care quality programme: creating change through the use of quality improvement collaboratives. Health Policy. 2022;126:749–54. Article  PubMed  Google Scholar  * USFDA. Considerations for

the Use of Real-World Data and Real-World Evidence To Support Regulatory Decision-Making for Drug and Biological Products

https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-use-real-world-data-and-real-world-evidence-support-regulatory-decision-making-drug last accessed:

24/04/2024. * Roy A. Key Findings from the 2022 FREOPP World Index of Healthcare Innovation. available from:

https://freopp.org/key-findings-from-the-2022-world-index-of-healthcare-innovation-e2a772f55b92; last accessed: 24/04/2024. * Tikkanen R, Osborn R,EM, Djordjevic A, Wharton GA. International

Health Care System Profiles - Switzerland. _The Commonwealth Fund_ 2020; available at: https://www.commonwealthfund.org/international-health-policy-center/countries/switzerland last

accessed: 24/04/2024. * De Pietro C, Camenzind P, Sturny I, Crivelli L, Edwards-Garavoglia S, Spranger A, et al. Switzerland: Health system review. Health Syst Trans. 2015;17:1–288. * Jörg

R, Zufferey J, Zumbrunnen O, Kaiser B, Essig S, Zwahlen M, et al. The Swiss health care atlas—relaunch in scale. Res Health Serv Reg. 2023;2:3. Article  PubMed  PubMed Central  Google

Scholar  * Mertins T, Nilius H, Boss R, Knuchel M, Signorell A, Huber CA, et al. Secondary prevention of venous thromboembolism: Predictors and outcomes of guideline adherence in a long-term

prospective cohort study. Front Cardiovasc Med. 2022;9:963528. Article  PubMed  PubMed Central  Google Scholar  * Reich O, Bachmann LM, Faes L, Böhni SC, Bittner M, Howell JP, et al.

Anti-VEGF treatment patterns and associated health care costs in Switzerland: findings using real-world claims data. Risk Manag Healthc Policy. 2015;8:55–62. Article  PubMed  PubMed Central

  Google Scholar  * Reich O, Schmid MK, Rapold R, Bachmann LM, Blozik E. Injections frequency and health care costs in patients treated with aflibercept compared to ranibizumab: new

real-life evidence from Switzerland. BMC Ophthalmol. 2017;17:234. Article  PubMed  PubMed Central  Google Scholar  * Schmid MK, Reich O, Blozik E, Faes L, Bodmer NS, Locher S, et al.

Outcomes and costs of Ranibizumab and Aflibercept treatment in a health-service research context. BMC Ophthalmol. 2018;18:64. Article  PubMed  PubMed Central  Google Scholar  * Schmid MK,

Reich O, Faes L, Boehni SC, Bittner M, Howell JP, et al. Comparison of outcomes and costs of ranibizumab and aflibercept treatment in real-life. PloS One. 2015;10:e0135050. Article  PubMed 

PubMed Central  Google Scholar  * Excellence NIfHaC. Ranibizumab for treating diabetic macular oedema - Technology appraisal guidance Reference number: TA274.

https://www.nice.org.uk/guidance/ta274 Last updated: 26 October 2023, last accessed 24/04/2024 * Cheung N, Cheung CMG, Talks SJ, Wong TY. Management of diabetic macular oedema: new insights

and global implications of DRCR protocol V. Eye (Lond). 2020;34:999–1002. Article  PubMed  Google Scholar  * Do DV, Moini H, Wykoff CC. Frequency and timing of antivascular endothelial

growth factor treatment for eyes with centre-involved diabetic macular oedema and good vision: Protocol V results in context. BMJ Open Ophthalmol. 2022;7:e000983. Article  PubMed  PubMed

Central  Google Scholar  * Rennie C, Lotery A, Payne J, Singh M, Ghanchi F. Suboptimal outcomes and treatment burden of anti-vascular endothelial growth factor treatment for diabetic macular

oedema in phakic patients. Eye (Lond). 2024;38:215–23. Article  PubMed  Google Scholar  * Flaxel CJ, Adelman RA, Bailey ST, Fawzi A, Lim JI, Vemulakonda GA, et al. Age-related macular

degeneration preferred practice pattern(R). Ophthalmology. 2020;127:P1–P65. Article  PubMed  Google Scholar  * Ross AH, Downey L, Devonport H, Gale RP, Kotagiri A, Mahmood S, et al.

Recommendations by a UK expert panel on an aflibercept treat-and-extend pathway for the treatment of neovascular age-related macular degeneration. Eye (Lond). 2020;34:1825–34. Article 

PubMed  Google Scholar  * Smith R. Dartmouth Atlas of Health Care. BMJ. 2011;342:d1756. Article  Google Scholar  * Mays N. Reducing unwarranted variations in healthcare in the English NHS.

BMJ. 2011;342:d1849. Article  PubMed  Google Scholar  * Westert GP, Faber M. Commentary: the Dutch approach to unwarranted medical practice variation. BMJ. 2011;342:d1429. Article  PubMed 

Google Scholar  * Fassler M, Jobges S, Biller-Andorno N. Bonus agreements of senior physicians in Switzerland - A qualitative interview study. Z Evid Fortbild Qual Gesundhwes.

2020;158-9:39–46. Article  Google Scholar  Download references FUNDING Open access funding provided by University of Luzern. AUTHOR INFORMATION AUTHORS AND AFFILIATIONS * Eye Clinic, Lucerne

Cantonal Hospital LUKS, Lucerne 16, Switzerland Martin K. Schmid & Michael A. Thiel * University of Zurich, Zurich, Switzerland Martin K. Schmid, Lucas M. Bachmann & Michael A.

Thiel * Department of Health Sciences and Medicine, University of Lucerne, Luzern, Switzerland Martin K. Schmid & Stefan Boes * Institute of Ophthalmology, University College London,

London, UK Dawn A. Sim * Genentech South San Francisco, South San Francisco, CA, USA Dawn A. Sim * Jules Gonin Eye Hospital, University of Lausanne, Lausanne, Switzerland Thomas J.

Wolfensberger * Medignition AG, Zurich, Switzerland Lucas M. Bachmann * Vista Eye Clinic Binningen, Binningen, Switzerland Katja Hatz * Faculty of Medicine, University of Basel, Basel,

Switzerland Katja Hatz Authors * Martin K. Schmid View author publications You can also search for this author inPubMed Google Scholar * Dawn A. Sim View author publications You can also

search for this author inPubMed Google Scholar * Stefan Boes View author publications You can also search for this author inPubMed Google Scholar * Thomas J. Wolfensberger View author

publications You can also search for this author inPubMed Google Scholar * Lucas M. Bachmann View author publications You can also search for this author inPubMed Google Scholar * Katja Hatz

View author publications You can also search for this author inPubMed Google Scholar * Michael A. Thiel View author publications You can also search for this author inPubMed Google Scholar

CONTRIBUTIONS MKS, MAT, and LMB initiated the project. All authors contributed to the drafting and approved the final version. CORRESPONDING AUTHOR Correspondence to Martin K. Schmid. ETHICS

DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. ADDITIONAL INFORMATION PUBLISHER’S NOTE Springer Nature remains neutral with regard to jurisdictional claims in

published maps and institutional affiliations. RIGHTS AND PERMISSIONS OPEN ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use,

sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative

Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated

otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds

the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Reprints and

permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Schmid, M.K., Sim, D.A., Boes, S. _et al._ Between Scylla and Charybdis?—Health insurance claims-data to monitor quality of service delivery

in ophthalmology. _Eye_ 38, 3412–3415 (2024). https://doi.org/10.1038/s41433-024-03333-5 Download citation * Received: 05 May 2024 * Revised: 21 August 2024 * Accepted: 05 September 2024 *

Published: 18 September 2024 * Issue Date: December 2024 * DOI: https://doi.org/10.1038/s41433-024-03333-5 SHARE THIS ARTICLE Anyone you share the following link with will be able to read

this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative