Session 11 Drug and LTD Issues. Denis Garand, FCIA, FSA June 21, 2017 With contributions from Ken Fraser and Donna Swiderek

Transcription

1 Session 11 Drug and LTD Issues Denis Garand, FCIA, FSA June 21, 2017 With contributions from Ken Fraser and Donna Swiderek

2 CIA translated by voice mail = Canadian Institute of Actual Worries Issues in LTD The risk is changing in all its components, incidence, terminations, integration, rehabilitation Do you understand the risk Issues in Rx coverage

3 How do you review LTD results? Do you have any bench marks? Is LTD really long? 60% of claim value in fist 2 years 20% of claim value for claims beyond 8 years! What rate per thousand of claims with duration beyond 4 years? What percent of those are integrated with C/QPP?

4 Considerations in pricing and renewals Data robustness Actuarial tables Company target markets, distribution Company underwriting skill Company investment function Company claims and rehab management practices Profit goals considering risk based capital and tax management Margins in measurement of liabilities Product offered and contractual terms A pricing actuary should have a good knowledge of all of these as they impact actual results.

5 Data LTD a large risk and a critical issue WHY IS THE MANAGEMENT OF DATA FILES NOT A CRITICAL PRIORITY? Retain claims record for minimum of 15 years Date of all transactions on the claimant Maintain accurate information on claimant especially on data fields that are used for pricing or reserving. Avoid work arounds on fields Ensure that the same data base is used by valuation, pricing, administration and accounting.

6 Data Accuracy Periodic review summary of key database fields, assess for reasonableness. Insure downloads from data base are accurate. Reconciliation of extracted data to ensure correct fields are downloaded in merges In our studies we have picked up errors from data created by merging two data sources, when the original data was correct. These issues can impact analysis WEAK DATA IS COSTLY Annual Actual/Expected (A/E) studies on key assumptions

7 Current approaches to pricing and renewals for LTD Pricing Using assumptions on incidence, termination rates, claims and other expenses, investment returns, expected profit, integration, etc. Reviewing portfolio for areas of gain and loss based on current reserving assumptions. Present value of cash flow with discounting of the most recent DLRS Renewals Combining experience, based on reserving assumptions with pricing assumptions and various formulas of credibility, depending on size. DO RESERVING ASSUMPTIONS IMPACT THESE METHODS?

8 Issues with reserving assumptions in assessing emerging results In ALL the cases we review, claims results in the first two to four years of a group/block are overstated. What is not captured in reserving assumptions Delays in claims approvals Partial/temporary return to work Reduction for All Sources limit Delays in C/QPP integration and non recognition of recapture of payment Large benefit amount to a young person will attract a LARGE Liability

9 Towards a clear view of results Liabilities should be set: Without pads Integrated with C/QPP, with expected recovery Estimates of reduced liability for rehab period Claims beyond 2 years duration adjusted for real likelihood of recovery if not integrated with C/QPP Need to review by year of incurral and understand overstatement of liabilities.

10 Critical benchmarks to use in review of results By incurred loss year Actual open claims For each policy year cohort, the incidence of claims still open after 4 years of disability Of those open after 3 years, % of that are integrated with C/QPP Discount of Cash flow and DLR Average claim per exposure, by age and gender, in $ and in Benefit amount. Versus expected if available to measure trends and shifts

11 How are you to make an informed underwriting decision? Case Study, 2+ million LTD premium, Ratios compared to premium Cash paid Claims Ratio Actual PY PY 1 PY 2 PY 3 PY 4 PY 5 PY6 Liabilities LR Actual LR Report PY 1 18% 13% 10% 9% 9% 7% 68% 134% 79% PY 2 18% 19% 4% 4% 3% 17% 65% 102% PY 3 14% 25% 19% 11% 44% 113% 59% PY 4 15% 20% 12% 57% 103% 151% PY 5 8% 13% 65% 87% 110% PY 6 8% 101% 109% 75% Total 102% 96% PY=Policy Year, LR= Loss ratio,

12 Using Benchmarks and incurred year analysis Benchmarks let you know if there are large deviations A/E incidence was lower then 100% A/E open at duration 3 years was lower then 80% CPP integration lower then expected, one long duration claim of high value had no CPP, it recovered in policy year 7, reducing PY1 claims ratio to 80% in incurred basis Also compared A/E on termination, case had very high terminations Incurred year analysis help redirect conversation Total expense of rehab, claims adjudication also will have an impact on long term results

13 Results by carrier A question we should all ask is there variability by carrier? Claims management, rehabilitation, back to work, integration practices Understand the capabilities of the carrier? A benchmark is an average, is there a logical reason to vary from the benchmark? Can you quantify expected variation => more relevant benchmark

14 Conclusions Do companies really understand their LTD block? LTD is a very important benefit. Why the inattention! Pricing must consider the actual cash flow Primarily claims but also investment income, expenses and premiums Pricing has to consider the actual skills of the insurer Skills = systems, capabilities and staff resources. Results will be different by carrier

15 Issues in Rx the cost of Rx ever increasing, biometrics taking a bigger share. High Amount Pooling, this has a an impact on future of benefits and post retirement benefits. (Are actuaries adjusting factors for these in post retirement liablities? Will constant increase result in reduction of health benefits? Are there other ways to approach this by industry/govt/research

16 Use of biometrics Canada, less then 65 In % of cost, 0.9% of incidence In % of cost, 0.5% of incidence Variation by region, in % Atlantic, 27.4% QC,26.8% in ON and 20% in SK/MB From 2008 to 2016 monthly drug cost went from $75 to $104, a 4.1% annual increase. Normal drugs went from $67 to $75, a 1.6% annual increase Biometrics went from $7 to $28, a 16.2% annual increase Source Telus Health

17 Changing cost of Rx 2005, 14 drugs cost > $ 10,000 a year. 2015, 124 drugs cost > $10,000 a year, including 20 >$50,000 and 45 >$20, 000 Drugs >$50,000, 1% of total cost in 2005, and 7.4% in Source: Jean Lachaine, professeur titulaire de la Faculté de pharmacie de l Université de Montréal

18 Internationally Canada has the second highest cost of Rx Example recent Globe and Mail article says it cost $500,000 a year to treat AIDS patient, in Kenya cost is USD 435 a year.

19 Potential solutions It is political! Someone has to argue the macro economic value of a different drug scheme..who WILL IT BE Plan design? Something else.