The Value of Healthcare Information Exchange and Interoperability in New York State

Transcription

1 The Value of Healthcare Information Exchange and Interoperability in New York State Prepared by the Center for Information Technology Leadership August 2005 The United Hospital Fund of New York (UHF) engaged the Center for Information Technology Leadership (CITL) to project the value of healthcare information exchange and interoperability (HIEI) for the state of New York. Based on a model and analysis developed for the United States, CITL used New York-specific inputs to determine the benefits and costs of systems that enable standardized information exchange between New York state healthcare stakeholders. Inputs were provided by UHF, with assistance from the Greater New York Hospital Association and Empire Blue Cross Blue Shield, or pro-rated based on the population of New York. This document summarizes the model s approach to each of its benefit and cost components, and reports projections of HIEI value for New York State. The model considers transactions between providers (hospitals and medical group practices) and the stakeholders with whom they most commonly exchange information: other providers, independent laboratories and radiology centers, pharmacies, payers, and public health departments. It quantifies the value of the entire system moving from today s prevailing phone and mail communications to an idealized state of full computer-to-computer, standardized data exchange with minimal human involvement. CITL quantified potential sources of benefit and cost for each relationship from evidence in the published literature, market research, and estimates by an expert panel. Once fully implemented, standardized HIEI could yield net value of $4.54 billon annually in New York. In addition, the state could net $12.4 billion in savings over an initial 10-year implementation period. The largest value accrues to providers, laboratories, and payers. Individual provider organizations breakeven at different points during the 10-year implementation period; while large medical offices see positive return in the first year, small hospitals do not breakeven until year 6. All projections are reported in 2003 US dollars and to three significant digits. Analytic Framework CITL projected the value of electronic data transactions flowing from clinical encounters between providers (defined as hospitals and medical group practices) and other providers, and between providers and five key stakeholders with which they exchange information most commonly: independent laboratories and radiology centers, pharmacies, payers, and public health departments. The flow of transactions is represented by the arrows in Figure 1: 1

2 Figure 1 Overview of Provider-Stakeholder Relationships Public Health Other Providers Radiology Provider Pharmacy Payer Laboratory CITL defines four levels of interoperability (Figure 2). The model calculates the value of moving from Level 1 to Level 3 and Level 4. Level 2, machine-transportable data exchange, is included in Figure 2 for reference but was not included in the New York state analysis. Figure 2 Levels of Healthcare Information Exchange and Interoperability Level Description Examples 1 Non-electronic data exchange Mail, telephone 2 Machine-transportable data Standard fax, ed pictures 3 Machine-organizable data of free text, PC-based exchange of files in incompatible/proprietary file formats, HL-7 messages 4 Machine-interpretable data Interoperable data exchange with standardized message formats and content, automated exchange of coded results from an external lab into a provider s electronic health record, automated exchange of a patient s problem list The analysis projects organization-level value to typical medical offices and hospitals in the state. The medical office practice size categories are 5, 10, and 25 physicians for small, medium, and large groups respectively. The hospital size categories for small, medium, large, and jumbo hospitals are 49 beds or fewer, beds, beds, and 400+ beds respectively. The model projects annual, steady-state value after full implementation, and value over a 10-year implementation period. It assumes 20% of users implement HIEI each year and each incurs acquisition costs in the first year and annual costs beginning the same year. Each user reaps 50% of total HIEI benefit in its first year, and benefit climbs 10% per year until it reaches 100% in the 2

3 sixth year. By year 10, all participants reach 100% of their benefit potential. Table 1 includes a benefit realization schedule. For further information about the model and CITL methods, please refer to reports of the original model that forms the foundation for this analysis. 1, 2 Benefits of HIEI Quantitative and qualitative benefits of interoperability are described below. For year-by-year results, and proportions of benefits accruing to different organizations, please refer to Table 2. Benefit of interoperability between outpatient providers and independent laboratories Both clinicians in free-standing practices and those that are hospital-based use external laboratories. Interoperability between these organizations would enable computer-assisted reduction of redundant tests, and it would reduce delays and costs associated with paper-based ordering and reporting of results. These savings would produce the following annual benefit to New York: Level 3 $9.69 billion $1.49 billion Level 4 $16.3 billion $2.50 billion Additional benefits of provider-laboratory connectivity not quantified in the model include better clinician access to patients longitudinal test results, elimination of errors associated with verbally reporting results, optimization of ordering patterns by making test cost information readily available to clinicians, and more convenient testing for patients. Benefit of interoperability between outpatient providers and radiology centers Most imaging procedures ordered by office-based clinicians, and some ordered by those in hospital-based ambulatory practices, are performed in external radiology centers. Connectivity between these organizations would reduce redundant tests and would save time and costs associated with paper- and film-based processes. These improved efficiencies would save the state: Level 3 $5.86 billion $902 million Level 4 $9.79 billion $1.51 billion Interoperability could also improve ordering by giving radiologists access to relevant clinical information, thereby enabling them to recommend optimal testing; improve patient safety by alerting both the provider and the radiologist to test contraindications; facilitate coordination of 3

4 care and help prevent errors of omission by enabling automated reminders when follow-up studies are indicated; and lessen adverse environmental impacts by reducing the use of chemicals and paper in film processing. Benefit of provider-pharmacy interoperability Interoperability between outpatient providers and pharmacies would reduce the number of medication-related phone calls for both clinicians and pharmacists. As a result, the state would save: Level 3 $1.41 billion $217 million Level 4 $1.44 billion $221 million A large portion of benefit results from the reduction of illegible prescriptions and associated phone calls, which occurs before Level 3. Once Level 3 interoperability is attained, the only additional gain in moving to Level 4 is time saved on an additional 1.4% of calls between providers and pharmacies. Thus, the gain in value between Level 3 and Level 4 is modest. Outpatient provider-pharmacy interoperability would also improve clinical care by facilitating the formation of complete medication lists, thereby reducing duplicate therapy, drug interactions and other adverse drug events, and medication abuse. It could enable automated refill alerts, offer clinicians easy access to information about whether patients fill prescriptions, and complete insurance forms required for some medications. In addition, it could help identify affected patients in the event of drug recalls, uncover new side effects, and improve formulary management. Benefit of provider-provider interoperability Patients often see multiple clinicians, and their medical records are frequently scattered across several offices and hospitals. As a result, provider-provider connectivity would save time associated with handling chart requests and referrals. The model assumes that all needed charts are requested and projects the following benefits from these time savings: Level 3 $3.99 billion $614 million Level 4 $6.53 billion $1.01 billion Moreover, connectivity would reduce fragmentation of care from scattered records and improve referral processes by giving clinicians the patient-specific information they need to consult effectively on a case. Benefit of provider-payer interoperability 4

5 Payers and providers exchange administrative data in order to document services delivered and to ensure that providers are reimbursed according to contracted rates. Provider-payer transactions currently enjoy a relatively high degree of standardization, largely because of HIPAA (Health Insurance Portability and Accountability Act of 1996). HIPAA does not allow Level 3 connectivity. While some transactions are highly automated as a result of HIPAA, others are not, particularly in smaller organizations. Provider-payer connectivity would further improve current practices by making the exchange of seven categories of administrative data (eligibility inquiry and response, claims submission, claims attachments, claims status inquiry, remittance advices, referrals and pre-authorizations, and coordination of benefits) more efficient. This would save the state: Level 3 Not Allowed Under HIPAA Level 4 $6.78 billion $1.04 billion Benefit of provider-public health interoperability The public health system is a network of local, state, and federal agencies that pursue a wide array of population health objectives. Provider connectivity to the U.S. public health system would make reporting of vital statistics and cases of certain diseases more efficient and complete, saving the state: Level 3 $54.6 million $8.40 million Level 4 $99.3 million $15.3 million Though not quantifiable, the most significant impact of public health interoperability will almost certainly derive from earlier recognition of emerging disease outbreaks and bio-surveillance, as it becomes far easier to identify warning signs and trends by aggregating data from many sources. Costs of HIEI The analysis considers costs of internal clinical and administrative systems for providers, provider interfaces to stakeholders, and stakeholder interfaces to providers. Stakeholder system costs are not included. Table 3 contains cost detail. Medical office and hospital system costs Level 3 and Level 4 HIEI require providers to have broad and mature clinical information systems. The model projects costs for systems incorporating functions specified by the Institute of Medicine as minimally acceptable for electronic health records (EHRs): health information, patient support, results management, administrative processes, order entry and management, reporting and population health, decision support, and electronic communication. 3 The model 5

6 assumes all medical offices and hospitals in the state acquire and maintain new systems. Statewide system cost projections are as follows: Levels 3 & 4 $20.2 billion $1.17 billion Provider interface costs CITL estimates the cost of developing each HIEI-capable interface to be $50,000 per hospital and $20,000 per medical office, and annual maintenance to be 17.5% of those costs, beginning in year one. Level 3 requires a unique interface to each external organization, from eight to twenty interfaces per provider, depending on the provider s size. At Level 4, all message formats and vocabularies are standardized; therefore, each provider needs only five interfaces, one to each type of external organization. The model applies these costs to New York hospital and office statistics and projects the following costs: Level 3 $12.0 billion $874 million Level 4 $7.39 billion $527 million Cost of interfaces for pharmacies, laboratories, radiology centers, payer organizations, and public health departments Each pharmacy, laboratory, and radiology center needs one interface to provider EHR systems. Public health departments would likely maintain separate interfaces for hospitals and offices, and the model assumes two per local office. Costs are $50,000 per interface for development and 17.5% annual maintenance beginning in year one. The model applied these figures to the number of New York laboratories, radiology centers, pharmacies, and local health departments. Payer costs, only included at Level 4, are based on the HIPAA Final Impact Analysis. 4 The model relies on the Analysis s total payer cost estimates to comply with HIPAA standard transactions. The difference in cost between Level 3 and Level 4 is a result of the inclusion of Payer costs in Level 4. HIPPA assigns all costs to start-up with no on-going annual cost. Statewide costs are as follows: Level 3 $688 million $50.2 million Level 4 $927 million $50.2 million Net Value of HIEI 6

7 Combining the benefits and costs quantified above, the net value of interoperability at each level is: Level 3 Benefit $21.0 billion $3.23 billion Cost $32.9 billion $2.10 billion Net Value -$11.9 billion $1.13 billion Level 4 Benefit $40.9 billion $6.29 billion Cost $28.5 billion $1.75 billion Net Value $12.4 billion $4.54 billion Level 4 annual, steady-state net value for all participants is summarized in Figure 3 (see Table 4 for detail). The sum of the value shown is $5.72 billion. Providers annual system maintenance costs of $1.17 billion were not allocated to the illustrated transactions. When they are subtracted, the total net value to providers is $1.80 billion and the statewide net value is $4.54 billion. Figure 3 Annual Steady State Net Value at Level 4 in millions All medical offices and hospitals achieve positive net value at the end of 10 years, but they break even at different points (Figure 4, Tables 5 and 6). Medium and large medical offices reach positive net value in the first year and attain $19.5 and $118 million dollars in net value respectively over 10 years. Small hospitals are the slowest, not reaching positive net value until year 6 and reaping $1.23 million over 10 years. Information about the costs and benefits of each provider relationship can be found in Table 7. 7

8 Figure 4 - Cumulative Net Value per Medical Office or Hospital at Level 4 during Implementation $120,000 Small Office Medium Office Large Office Small Hospital Medium Hospital Large Hospital Jumbo Hospital $100,000 $80,000 $60,000 $40,000 $20,000 $- $(20,000) Years in millions Sensitivity Analyses Implementation period net value is most sensitive to office system costs, test costs, and interface costs. The tornado diagram in Figure 5 reflects the percent by which the Level 4 HIEI 10-year net financial value would change if factors were increased and decreased by 25%. For example, if medical office system costs were decreased by 25%, the total 10-year net value would increase 36% from $12.4 billion to $16.8 billion, whereas increasing the cost by 25% lowers 10-year net value 36% to $7.88 billion. As a result, any effort that results in lower medical office system costs will lead to the greatest increase in value. Steady state value is sensitive to substantially the same factors, except for office system costs, which have a much smaller influence on value since acquisition costs were incurred in the implementation period. Sensitivity analysis results are included in Tables 8 and 9. 8

9 Figure 5 - Sensitivity of Level 4 Net Value during 10-Year Implementation to Changes in Key Factors Medical office system cost Laboratory tests PMPY Radiology tests PMPY Provider interface cost Number of chart requests % esoteric tests (lab) Pages per chart Claims attachment volume Avg test cost (rad) Hospital system cost Percent of studies redundant (rad) % esoteric tests (rad) -40% -30% -20% -10% 0% 10% 20% 30% 40% % change in 10-year net value from +/- 25% change in variable value New York versus National Value The steady-state net value of Level 4 HIEI, $4.54 billion, is approximately 3.3% of $139 billion in total 2003 health care expenditures in New York. 5 In the national analysis, the projected annual net value of $77.8 billion is approximately 4.7% of annual expenditures of $1.66 trillion in The smaller ratio in New York State is largely due to three factors. First, New York has more physicians per capita than the US, making office system costs relatively higher. Second, New York has lower costs for radiology tests and, as a result, avoided tests do not incur as much benefit. Lastly, New York already handles more provider-payer transactions electronically than the rest of the country. For example, 100% of Medicaid hospital claims are processed electronically in New York, compared to 89% in the US. Again, there is less opportunity to derive value from HIEI in comparison to the US as a whole. Limitations With little real world experience with HIEI or its impact, quantitative evidence about its value is limited. The analysis incorporates the best evidence available, combining estimates from experts and a small number of studies. The figures above represent a financial analysis that does not impute a dollar value to important improvements derived from HIEI that yield additional clinical or organizational value. CITL suspects that the clinical payoff in improved patient safety and quality of care could dwarf the 9

10 financial benefits projected from this model, which are derived from redundancies that are avoided and administrative time saved. Additionally, the model considers only transactions between providers and five other entities, ignoring benefits accruing from transactions among many other participants that would be supported by HIEI. Further, it does not assign a dollar value to many other benefits that will accrue such as the value of HIEI-capable systems within an enterprise. As a result, the quantitative analysis cannot be considered comprehensive. The national model uses billed test cost for laboratory and radiology procedures. Using reimbursed test costs in the New York analysis may overestimate or underestimate various parts of the cost-benefit calculations; the ultimate impact on the overall results is not predictable. The analysis ignores care that crosses state lines, another area where impact is unpredictable. Just as the model underestimates the value of some important benefits because of lack of available data, it underestimates some costs. While the analysis calculates costs for providers to acquire HIEI-capable systems, it does not account fully for similar costs incurred by laboratories, radiology centers, pharmacies, payers, and public health departments. Also, it does not estimate the cost of developing and implementing state-wide standards or adopting national standards that will be essential for achieving Level 4 HIEI, or for the cost of major workflow disruptions during systems implementation. If employees are redeployed, the financial benefits projected from time savings may be realized as improved productivity or service quality, rather than pocketed dollar savings, and the model did not distinguish between these endpoints. Conclusion The results of this analysis indicate that statewide implementation of health care information exchange and interoperability is a good investment. Fully standardized HIEI could yield New York State a net value of $4.54 billion annually, or approximately 3.3% of total 2003 state health care expenditures. Even while stakeholders incur costs of installing systems during a 10-year implementation period, Level 4 interoperability is financially positive. Additionally, the clinical impact of HIEI for which quantitative estimates cannot yet be made would likely add further value. References 1 Pan E, Johnston D, Walker J, Adler-Milstein J, Bates DW, Middleton B. The Value of Healthcare Information Exchange and Interoperability. Chicago: Health Information Management and Systems Society, Walker J, Pan E, Johnston D, Adler-Milstein J, Bates DW, Middleton B. The Value of Health Care Information Exchange and Interoperability. Health Affairs. January 19, Committee on Data Standards for Patient Safety. Key Capabilities of an Electronic Health Record System. Institute of Medicine. National Academies of Sciences Standards for Electronic Transactions and Code Sets: Final Impact Analysis. Federal Register, August 17, CMS. United States Personal Health Care Expenditures (PHCE), All Payers Available at 10

11 CITL estimated 2003 health expenditures in NY based on 2000 PHCE, $94,769m. In the US, the 2000 PHCE was $1,136,115 million and 2003 total expenditures are estimated to be $1,661,000 million. CITL applied the US ratio to $94,769 million to get an estimate of $138,600 million for 2003 total NY healthcare expenditures. 6 Center for Medicare & Medicaid Services: Table 3: National Health Expenditures Aggregate and per Capita Amounts, Percent Distribution and Average Annual Percent Change by Source of Funds: Selected Calendar Years , 27 September Available at 11