Actuarial LTC Claims Management Session #17 Monday, March 4 3:30pm Joe Furlong, RGA Joan Stear, PennTreaty Dawn Helwig, Milliman 1
Actuarial Measurement The Continuum of Claim Management Accept Doctor s certification and Plan of Care (PoC) Use service that arranges discounts Send nurse to do care assessment Set standards for PoC, based on diagnosis Provide wellness benefits 0% Claim Management 100% Claim Management Review cert for reasonability Review PoC for reasonability Provide care coordinator to help with creating PoC Make services available (fall prevention, etc.) 2 2
Claims Management Claims Management Strategies Joe Furlong Associate Director, Claims RGA Reinsurance Company 3
Claims Management Strategies Claims Intake Process Fact Gathering Care Setting Reason for Care Setting Expectations Explain Policy Requirements Benefit Triggers Policy Benefit Limitsi Potential Recovery 4
Claims Management Strategies Plans of Care External Internal Length of Stay Benchmarks The Challenge of Co-Morbidities Diabetes Depression Arthritis 5
Claims Management Strategies Providers of Care Facility Based Care Nursing Homes Assisted Care Facilities Home and Community Based Care Certifying Independent Caregivers Adult Day Care Respite Care 6
Claims Management Strategies Managed Care Networks Database abase of Providers Licensing Care Provided May Shorten Approval Time Provider Discounts Pre-arranged savings 7
Claims Management Strategies Quality Assurance Random Auditing Over Authorization Reviews Initial Claim Decision Reviews Payments Denials as 8
Claims Management Strategies Managing Challenging Policy Provisions Restoration o of Benefits e Care Based Trigger Older Policies Recovery Based Trigger Newer Policies Alternate Plans of Care Home Modifications Alternative Care Arrangements Going Out of Contract 9
Claims Management Strategies Situations Which May Warrant Additional Review Younger Claimants Could mean shorter claims or longer claims Cash Benefits Possible incentive to remain on claim Recovery No longer benefit eligible 10
Claims Management Strategies Fraud Red Flags Investigations S.I.U. IMEs I.M.E.s Surveillance 11
Claims Management Measures of Success Joan Stear SVP Claims Management Penn Treaty Network America Insurance Company 12
Measures to Success If the elevator to success is out, take the stairs one step at a time. Author Unknown 13
Measures of Success Questions to Consider How Do You Define e Success? Operational Transactions Policyholder Satisfaction Thresholds What level are you willing to accept as an acceptable threshold? How Will You Measure it? Claims Audit Customer Satisfaction Survey Other Opportunities that may not Involve Audit, but Provide Valuable Feedback 14
Measures of Success How Will You Evaluate it? Establish s Transactional a Metrics and Monitor o Complaint and Appeal Tracking Policyholder Satisfaction Surveys 15
Measures to Success Getting From Point A to Point B Identify Areas of Risk Evaluate Tools Currently Used to Measure Effectiveness Revise or Create New Tools To Measure Implement Monitor Evaluate Effectiveness 16
Measures to Success Areas of Claim Risk Payment of Benefits e Eligibility Determinations Plans of Care Complex Products Restoration of Benefits Communication of Information 17
Measures of Success Claims Audit Transactional Audit How extensive do you want the audit to be?» Current Payment, Last 3 Payments Determinations Benefit Eligibility Provider Eligibility Restoration of Benefits Medical Necessity Determination Adherence to Claims Protocol Focused Audits Complex Products By Examiner 18
Measures to Success Plan of Care One of the Most Challenging Areas to Measure and Quantify Effectiveness Obstacles Co-morbidities Plan of Care Development is Based on Subjective Information Provided by the Policyholder. Evaluate Tools Utilized Face to Face Assessment vs. Medical Records Development of Assessment Tool Geared to Company Needs 19
Measures to Success Plan of Care Areas of Risk Type of a Claim Acuity of a Claim Plan of Care Development Establish Threshold Plan of Care Guidelines Benchmarks Measuring Effectiveness Audit of Plans Of Care Against Guidelines and Benchmarks Reduction of Claim Dollars Paid Based on Dx. 20
Measures to Success Restoration of Benefits Challenges Policy Language Ending a Claim Just Prior to Benefits Exhausting A New Approach Proactive vs. Reactive 21
Measures to Success Proactive Approach Development e e of 90% Benefits e Paid Report End of Care Assessment Validation of Assessment Findings with Policyholder by Phone Letters to Policyholder and Physician Noting Assessment Findings Measuring Effectiveness Claims Audit Monitoring of Appeals Received for ROB 22
Measures of Success Intake Process Area of Risk Accuracy of Information Provided Compliance with State Regulations Measuring Effectiveness Claims Audit Customer Satisfaction Survey 23
Measures to Success What is the End Result? Better e Claims Management age e Supported by Meeting Identified Thresholds Improved Customer Satisfaction Identification of Areas Requiring Retraining Improved Communication with Operational Departments Success is the sum of small efforts, repeated day in and day out. Robert Collier 24
Measures to Success And Remember.. The road to success is always under construction. Lily Tomlin 25
Claims Management Actuarial Measurement Dawn Helwig Consulting Actuary Milliman, Inc. 26
Actuarial Measurement Actuarial evaluation of new care management services An actuarial evaluation is essential Some changes in claim practices or new care management services could actually increase net costs Examples care coordinator could be more customer-focused and could recommend MORE care than would have been selected by claimant or could approve paying for some informal care, which would increase awareness and thereby increase claims 27
Actuarial Measurement Different services could affect different components of cost Claims management could change: Claims frequency Length of claim Utilization of services (# of days/week) Mix of services (NH/ALF/HHC) Cost of claim Claim expenses all of which affect product profits 28
Actuarial Measurement Traditional formal testing methods Randomized controlled trial two random groups established. One receives the test, the other doesn t. Risk profile matched control group create two risk-matched groups. One receives the test, the other doesn t. Population pre/post experiment look at PMPM claim results in population before test and after. 29
Actuarial Measurement Challenges of randomized control and risk profile-matching methods with LTC Many items drive LTC claim costs, including at least: Age and duration Gender BP/EP selected Marital status Underwriting method Geographic area Risk classification Etc. 30
Actuarial Measurement Challenges, cont. Therefore, creating a control group and a test group with identical risk characteristics is almost impossible with LTC! In addition, care management technique being added might take a long time to show any impact 31
Actuarial Measurement Population pre/post method Traditionally, compares claim cost before and after test is applied. Challenge is that this claim cost is expected to change, due to mix of population and aging. Modification: use detailed actual to expected testing to determine whether change has expected effect Basis of expecteds for A to E study must be robust enough to capture all material variances, so that a change in distribution does not affect the results Apply past, validated claim costs to actual postexperiment demographics to get expected frequencies, claim costs, etc. Compare to actual. 32
Actuarial Measurement A to E study Should show actual to expecteds separate by frequency, utilization and length of stay Ideally, should be done for facility separate from home care, and in total A to E study should be available with high degree of drill down capability A to E should also be performed on claim termination rates, at a high level (# of claims remaining after 3 months, 6 months, etc.) 33
Actuarial Measurement Length of time needed to evaluate For mature block of business,,probably need at least a year to see impact on incidence rates or on utilization rates Look at population size times expected frequency to get expected # of claims and use Longley-Cook credibility table to get number of claims needed for desired confidence How many years of experience does it take to get to that claims level? Length of stay impact could take longer, though earlier results could be seen by looking at claim continuance before and after 34
Actuarial Measurement For favorable impact, look for: Reductions in incidence ce rate A to E s and/or utilization rates and costs, or increases in claim termination rates Need to make sure total impact results in reduction in agregate A to E s Some claim initiatives might focus on certain diagnoses (falls, cognitive), so may need to be able to compare frequencies for that diagnosis, before and after 35
Actuarial Measurement Experiment that probably does not work Can t offer a claim management technique on an optional basis and expect to be able to measure # of claims in group from people who selected technique vs. those who didn t Those that select the technique might have been at greater risk than rest of the population Example offering a cognitive tool on an optional basis. Those at greatest cognitive risk are likely to be the ones who select. Alternately, those who select might be the healthier and more risk-adverse. 36
Actuarial Measurement Example wellness program early identification of claim Care coordinator avoids a later nursing home stay and starts person in HHC at earlier date NH frequencies will go down (but not initially) HHC frequencies will go up (immediately) NH lengths of stay may go down and HHC lengths of stay may go up (less severe NH conditions are the ones moved, but they might be more severe than the average HHC condition) Average HHC utilization may go up What is the impact on total claims costs? 37
Actuarial Measurement Hypothetical Example of Possible Impact of Care Management Nursing Home Home Health Care Unmanaged Well Managed Unmanaged Well Managed Frequency.0246.0184.0220.0264 # Months Disabled 20.1 22.1 18.4 20.1 Cost/Service $120 $120 $47 $50 % Using Services 100% 100% 66% 80% # Services/Month 30.4 30.4 27.8 29.6 Annual Claim Costs $1,804 $1,483 $349 $628 Total Cost: Unmanaged = $2,153 Managed = $2,111 2% savings 38
Actuarial Measurement Evaluating Changes from prior slide can only be measured in aggregate For wellness program example, measurement would have to take place over longer period of time, to catch the avoided NH stay A to E s on total block gives a way to look at the aggregate impact 39
Actuarial Measurement Complicating factor with short evaluation timeframe There is natural variance in the frequency, length of stay and utilization in any block of LTC business Average standard deviation for a block of LTC business: 15-20% per year for smaller blocks 5% for larger blocks Natural morbidity improvement? 40
Actuarial Measurement A statistical i variance in the year of measurement could lead to an erroneous conclusion May need to expand the testing timeframe to get number of claims for needed for credibility May need to drill down further any other factors that have changed and affected the results? May need to rely on specific observations of the claims department, based on individual claims (e.g., claimant A moved to HHC services cost 80% of what NH would have cost) 41
Actuarial Measurement Bottom line Measuring impact of claims initiatives takes time and could be difficult to do Would likely only be done for very extensive or costly claims initiative Use caution in building any assumed savings into pricing or gross premium valuation until they re confirmed 42