INCENTIVE OF DRGS MARTTI VIRTANEN NORDIC CASEMIX CONFERENCE

INCENTIVE OF DRG’S? MARTTI VIRTANEN NORDIC CASEMIX CONFERENCE 3. 6. 2010

DIAGNOSIS RELATED GROUPS • Grouping of patients/episodes of care based on – diagnoses, – interventions, – age, sex, mode of discharge (and length of stay) • Casemix-system – using routine data from healthcare (hospital) activity MARTTI VIRTANEN 3. 6. 2010

BASIC PROBLEM • Need to compare hospital activity in health economics – Different hospitals have different patients – fair comparison of cost needs to take in account this distribution (called casemix) by standardizing for the casemix • Basic reason for the variation is the nature of healthcare – different individuals, different problems , different cost • To control for the variation one need groups of reasonable size where the resource needs can be mathematically described and expected needs calculated • The variation within the groups must be as small as possible to reduce the error of the prognosis. • To be able to use the data for development of the care the groups need to clinically meaningful MARTTI VIRTANEN 3. 6. 2010

BASIS OF THE GROUPING? • Routine data – Possibility to use old data – Available criteria of data quality • Specialty based groups – Lacking homogeneity within the groups • Diagnosis based groups – Too many different diagnoses – Large variation within the groups because of interventions • Intervention based groups – Too many different interventions – No interventions for all patients • DRG model: – Diagnoses and interventions must be grouped and groups need to be combined by rules MARTTI VIRTANEN 3. 6. 2010

BASICS OF THE DRG-TYPE CASEMIX SYSTEMS • Clinically and economically meaningful groups based on existing information (routine data) – Each group has to be clinically and medically understandable collection of patients – Each group has to be economically homogenous i. e. have low variance within the group MARTTI VIRTANEN 3. 6. 2010

PRODUCTS OF HEALTHCARE • The goal is to help not to sell (products) HOWEVER • Comparison of individuals in different units (hospitals) is not reasonable • Group of similar patients can be compared • The group formed for comparison is in practice a product • Combination of a product (group of patients) with generally accepted resource need of that group creates system of products • A system of healthcare products creates possibility to compare the work even in units of clearly distinct character – Cost level – Quality of care MARTTI VIRTANEN 3. 6. 2010

USE OF DRG • Hospital activity economical follow-up and evaluation – Adjusting for patient quality (casemix) – The original purpose of DRG’s at Yale • Basis of reimbursement – Prospective payment – DRG’s in their variations are the most popular method of hospital reimbursement in the world (after no system at all) MARTTI VIRTANEN 3. 6. 2010

BASIC DISTINCTIONS • Surgical and conservative groups – “Operation Room” procedure? – “Surgical” intervention? • Inpatient – What is “in”? – Diminishing natural resource • Outpatient – Day-Surgery? ? • Limits of “day” • Definition of surgery – intervention performed by surgeon? – Other outpatients MARTTI VIRTANEN 3. 6. 2010

COST AND PRICE • The price of a product must cover the cost of its production – There is no other connection – In a for-profit organization the price has to cover also the surplus • The price of an intermediate product is a cost of the end product MARTTI VIRTANEN 3. 6. 2010

COST BY PATIENT (CBP/KPP) • DRG defines a group for each patient • To calculate the cost of each group one needs the costs of each patient in the group – Some costs are by nature by patient – ‘Direct cost’ • • • An surgery performed on a patient A radiological examination on a patient Laboratory test taken from the patient An aid or a prosthesis given to a patient Medication given to a patient – The activity needs facilities and services that are not direction associated with any single patient – ‘Indirect cost’ • The working hours and salaries of the staff are not directly connected with any single patient MARTTI VIRTANEN 3. 6. 2010

MODELS FOR HEALTHCARE REIMBURSEMENT Capitation Per day & per visit Fee for service DRG 11 MARTTI VIRTANEN 3. 6. 2010

INCENTIVE AND OPTIMIZATION • The producer must optimize prices and costs by product – Cost has already been cut – Increasing level of activity? – Avoiding unnecessary use of healthcare services by patients? • The definition of the product guides the optimization process – Which parts should and could be optimized and how? • The challenge: Product definition where optimization of single products automatically results in optimization of the total service 12 MARTTI VIRTANEN 3. 6. 2010

CAPITATION • The producer has the responsibility for the availability of the services of a defined population • The price is set by the size of the population • Age, disease prevalence and special needs of the population are usually taken in account by standardization • Producer strategy: Good care diminishes producer cost by keeping healthy individuals at home • Incentive: Keep the patient away 13 MARTTI VIRTANEN 3. 6. 2010

PER DAY & PER VISIT • Price for inpatient day / outpatient visit is set to cover mean expenses – True cost variation is extremely large – Intervention days are very expensive for the producer • Producer strategy: Take good care of the patient by long hospitals stays and many outpatient check-up’s. • Incentive: Keep the beds warm i. e. produce as many cheap hospital days and cheap outpatient visits as possible 14 MARTTI VIRTANEN 3. 6. 2010

FEE FOR SERVICE • Often combined with for ex. per day & per visit model • Cost of interventions is covered by intervention specific price • Producer strategy: Be sure not to miss an intervention the patient might benefit of • Incentive: Do as much as possible – adverse effects will be corrected with additional interventions (for additional fee!) 15 MARTTI VIRTANEN 3. 6. 2010

DRG – CURRENT MODEL • Price per contact with care based on the problem and the care given – Surgical/conservative – Each contact results in a bill (to community) • Interventions, inpatient days, outpatient visits, lab-tests, radiological examinations etc are intermediate products that define the cost of the product (a DRG). • Producer strategy: – The patients are given good care by allowing them to come to the hospital whenever needed. – The contacts with care arranged effectively so that the necessary package of interventions is performed as rapidly as possible. – The patient goes home after each package and returns to hospital for the next package. • Incentive: Only the procedures affecting the DRG assignment should be performed at each contact. An episode of care should be cut to as many contacts as possible. 16 MARTTI VIRTANEN 3. 6. 2010

DRG – ORIGINAL MODEL • Developed in USA in late 60’s and early 70’s – DRG’s applied for inpatient care only • Typical model of healthcare delivery was long hospital stay – ‘Treated patient’ – concept • At discharge the problem was usually solved or dealt with for the time being • Episode of care – An expensive system that did not benefit the patient • Creates conceptually meaningful products • Producer strategy: Unnecessary long inpatient episodes were to be shortened • Incentive: Replace long inpatient episodes with repeated outpatient visits – Effects as expected! 17 MARTTI VIRTANEN 3. 6. 2010

EPISODE OF CARE • Combination of all contacts with healthcare for a given health problem • Different types of episodes (and examples) – Acute short term episode • Infection – Acute episode with prolonged need for care • Cholecystitis • Many malignancies • Coronary heart disease – Chronic disease episode • Permanent problems without curative care – Diabetes – Rheumatoid arthritis – Metabolic syndrome 18 MARTTI VIRTANEN 3. 6. 2010

DEFINITION OF SUBEPISODES OF CARE FOR REIMBURSEMENT 2005 EPISODES – ENDING EPISODES 2004 2005 2006 Subepisode 2005 – birthday to birthday Subepisode 2005 – a year from the beginning of problem Episode 2005 – care ending 2005 Episode not ending 2005 Episode starts but does not end 2005 19 MARTTI VIRTANEN 3. 6. 2010

REPEATING EPISODES • Acute infection – chronic repeating infections – First episodes will be regarded as separate infections – Criteria for repetivenes • Similar problems – Headache – Convulsions – Acute vs. chronic coronary heart disease 20 MARTTI VIRTANEN 3. 6. 2010

SIMULTANEOUS EPISODES Problems existing at the same time • Diabetes, hypertension and obesity – Metabolic syndrome? • Diabetes and hip fracture – Truly two different problems Problem developing from previous problem • Adverse effect of care – Part of the basic problem? ? – Separate problem – who pays • Acute care and rehabilitation – Necessary to separate because: • Some patients do not need rehabilitation – For example those dying during acute caree • Some hospitals do not have rehabilitation 21 MARTTI VIRTANEN 3. 6. 2010

DETECTION OF EPISODES WITH CURRENT HEALTHCARE DATA? • Same problem? – Same main condition? • Dx code may change during the treatment process – Main condition of a contact may appear as secondary dx at another contact – Same DRG? • Surgical and conservative episodes? – Same MDC? – Same diagnoosiscaategory • Both too loose? – Restriction by time? • Can be used to support previous models • One episode may continue for long period and inactive parts may also be long – Treatment by as few contacts as possible is often optimal care! • More than 90% of cases are obviously easy to detect – Problematic for reimbursement 22 MARTTI VIRTANEN 3. 6. 2010

HEALTHCARE PURCHASER STRATEGY • Goal – High quality healthcare services for the population avoiding uncontrolled cost development • Strategy – The cost of the activity must be covered for maintenance and effective use of the existing infrastructure – The reimbursement system must give an incentive • to control cost development and • to support good quality of care – Reliable prediction of cost is crucial for both purchaser and producer economy 23 MARTTI VIRTANEN 3. 6. 2010

ALL WISHES • Transparency • Justice • Reliable budgeting 24 – What are we paying for? • As many details as possible? – Detailed fee for service » Information is lost in the details! • Broadly summarized data? – Capitation » Never mind the details – We pay for our population, others have to take care of their own! • Need for as detailed information as possible – When we pay we get fast, effective good quality service (? ? ) – Dependent on disease incidence • How can our community of 1500 inhabitants cope with triplets at 28 weeks of gestation? – Adjustment of risk in large enough administrative units • But, but – people in the city visit the hospital for every minor problem and our citizens from outskirts go there only when really necessary! MARTTI VIRTANEN 3. 6. 2010

EPISODEDRG • DRG episode – Starts • by first health care contact for new health problem or • by first health care contact after end of previous subepisode of an existing health problem – Ends • When the problem is considered solved or • when preset time limit ends the subepisode • At each contact one can deal with more than one episode 25 MARTTI VIRTANEN 3. 6. 2010

COMPONENTS OF AN DRG EPISODE • Specialist level DRG episode includes: Inpatient care Outpatient care Day-surgery Laboratory contacts Radiology contacts Serial therapies • Radiation therapy, physiotherapy, series of medication – A new episode caused by adverse effects of care? ? • Intensive care in another hospital • Unplanned treatment at another hospital? – Care of the problem after it has been (falsely) considered solved – – – • Specialist level DRG episode does not include • Outpatient and inpatient care combined 26 – Contacts with primary care – Planned care on higher or lower level of care – Episodes treated only as outpatient are usually less resource intensive and should be detected as separate groups – The concept ‘Day-surgery’ mostly disappears MARTTI VIRTANEN 3. 6. 2010

EPISODE DRG FOR REIMBURSEMENT • Prospective system – Episode can be reimbursed when it has been reliably detected by given criteria • Usually straight forward • For repeating problems clear criteria are crusial – Episode is detected at latest when it is closed • Producer closes an acute health problem episode when it is considered solved – Producer is responsible during guarantee period for additional cost of the same problem 27 MARTTI VIRTANEN 3. 6. 2010

QUALITY AND DRG • Responsibility for possible continuation of care of the same problem • Responsibility for care of adverse effects • Mortality by DRG episode • Feedback from other levels of care – Primary care Specialist level care – Local hospital District hospital University hospital • Adverse effects! • Patient satisfaction by DRG episode – Evaluation of the total process of care 28 MARTTI VIRTANEN 3. 6. 2010

DEVELOPMENT OF EPISODE DRG • All episodes of year X – Episodes ending in year X • The analysis must be based on 3 year data – Cost from year X-1 must be include in the patient cost – Episodes continuing to year X+1 must be excluded (also from costs) • Using available information the contacts are combined to episodes – Special algorithm is necessary • Exact cost counting by patient is a must in the developing environment 29 MARTTI VIRTANEN 3. 6. 2010

LOPPU Thank you for your patience Questions? ! 30 MARTTI VIRTANEN 3. 6. 2010