INDEXATION OF MEDICAL COSTS FOR SOUTH AFRICAN MEDICAL SCHEMES

Roseanne da Silva

IAAHS 2007IAA Health Section Colloquium

13th – 16th May 2007CTICC

www.iaahs2007.com

Objectives

• Index for private sector medical prices• Experienced by members of a particular

set of medical schemes• In respect of claims submitted for

reimbursement• Opportunities for expansion

Ideal index characteristics

• Representative• Objective• Consistent• Smooth• Ease of understanding (and calculation)

MPI

Inflation is defined as the persistent change in the general price level

The intention of the index is to track the changes in the price of medical services 

over time in an objective way.

Currently available measures

• Contribution increases– Per scheme– Report of the Registrar

• CPI– Medical care and health expenses

Medical Scheme Contributions

• Used as proxy for medical costs• But includes:

– Non-healthcare costs– Reserving requirements– Utilisation trends– Impacts of benefit changes– Impacts of demographic changes

Medical scheme interactionsInfluenced by:Claim costsAdministration costsUtilisationBenefitsDemographicsContrib. to reserves

MPI is measure of the change in the price of services provided by the medical service

provider

Index development

• Measuring performance / effects• Monitoring growth• Benchmarks• Identification of problem areas

Price indices• LASPEYRES INDEXIndex =

Where Pit = Price of item i (i = 1,.....,m) in period tPio = Price of item i (i = 1,.....,m) in the base periodQio = Quantity of item i purchased in the base

period

( )( )P QP Q

it io

io io

∑∑

Price indices• PAASCHE INDEXIndex =

Where Pit = Price of item i (i = 1,.....,m) in period tPio = Price of item i (i = 1,.....,m) in the base periodQit = Quantity of item i purchased in period t

( )( )P QP Q

it it

io it

∑∑

Calculation process

• Household surveys (quantities)• Price surveys• Development of representative baskets• Aggregation of baskets

– Items to include– Substitution

• Weight allocations (within and between)• Index calculation

Sources of bias

• Substitution• Outlet substitution• Quality change• New product

(Boskin Commission, 1995)

ILO Manual

• Use of formulae• Frequency and completeness of surveys• Sampling methods• Quality adjustment• Relevance to sub populations• Differing objectives

South African CPI

• CPI• Core• CPIx• Inflation targeting• Weights based on 2000 household survey• 17 groups incl. medical care and health

expenses

Medical care and health expenses

100%7.15%Total: Medical care and health expenses

7%0.47%Insurance

14%0.98%Contributions to medical aid funds

2%0.16%Therapeutic appliances and equipment

37%2.61%Medical and pharmaceutical products

9%0.64%Hospital, nursing-home fees and fee for related services

32%2.29%Doctors fees, nurses fees and fee for related services

% of Medical Component of

CPI basket

% weighting

CPI

Chart of Medical Component of CPI and CPI (Base = 1990) for January 1990 to November 2005

0

100

200

300

400

500

600

700

Jan-9

0Ja

n-91

Jan-9

2Ja

n-93

Jan-9

4Ja

n-95

Jan-9

6Ja

n-97

Jan-9

8Ja

n-99

Jan-0

0Ja

n-01

Jan-0

2Ja

n-03

Jan-0

4Ja

n-05

Medical Component of CPICPI

Pilot study

• Medical schemes administered by sponsor• Service dates:1 Jan 2001 to 30 Nov 2005• Claim extracts:

• Hospital;• Medicine; and• Other (all remaining services providers and claim

types).

MPI

• Private sector measure• Calculated from medical scheme data• Claim amounts (not tariff)• Extensive baskets of products and

services• All items measured monthly• Adjustments for changing practice

Data issues

• Tariff coding– NHRPL– NAPPI

• Internal codes• Hospital consumables• Frequency definition

Basket Construction

57%63%52%52%48%Radiology51%49%50%47%50%Procedural - Specialist71%70%70%69%69%Pathology64%61%63%64%66%Optometry76%74%74%73%77%Obstetrics & Gynae80%77%76%78%76%General Practitioners54%58%61%65%55%Dentistry72%73%72%72%74%Consulting Specialist82%82%82%81%81%Anaesthetist69%69%69%69%69%Medicine Acute58%58%58%58%58%Medicine Chronic71%73%56%74%79%Hospitals20052004200320022001Basket

MPI calculation

• At per category level and overall• h(n) = hits weighting used in month n• p(n) = price in month n• M(n) = [sumproduct(h(n),p(n))] / sum (h(n))• Index(n) = [M(n) / M(n-1)] * 100

Linking methodology

• In transition month:– A = Weights(1) * cost per hit– B = Weights(2) * cost per hit

• Index in transition month = A/prior month• Index in transition month + 1 =

following month/B

Weightings

100%100%100%100%100%Totals

10%11%11%11%11%Other

26%26%25%25%26%Medicine

29%28%30%31%30%Hospitals

24%25%25%25%25%Specialists

10%10%8%8%8%General Practitioners

20012002200320042005MPI Components

Index weight

Comparison to CPI

7%Not includedInsurance14%Not included

Contributions to medical aid funds

37%36%Medicines9%9%Hospitals

13%Other providers28%Specialists

33%

14%General Practitioners

Medical component

of CPIMPIAt November 2004

MPI compared to CPI and the Medical component of CPI

for January 2001 to November 2005

90

100

110

120

130

140

150

Jan-

01Fe

b-01

Mar

-01

Apr

-01

May

-01

Jun-

01Ju

l-01

Aug

-01

Sep

-01

Oct

-01

Nov

-01

Dec

-01

Jan-

02Fe

b-02

Mar

-02

Apr

-02

May

-02

Jun-

02Ju

l-02

Aug

-02

Sep

-02

Oct

-02

Nov

-02

Dec

-02

Jan-

03Fe

b-03

Mar

-03

Apr

-03

May

-03

Jun-

03Ju

l-03

Aug

-03

Sep

-03

Oct

-03

Nov

-03

Dec

-03

Jan-

04Fe

b-04

Mar

-04

Apr

-04

May

-04

Jun-

04Ju

l-04

Aug

-04

Sep

-04

Oct

-04

Nov

-04

Dec

-04

Jan-

05Fe

b-05

Mar

-05

Apr

-05

May

-05

Jun-

05Ju

l-05

Aug

-05

Sep

-05

Oct

-05

Nov

-05

inde

x

MPI

Medical Component of CPI

CPI

MPI:5 major categories

80

100

120

140

160

180

Jan-

01Fe

b-01

Mar

-01

Apr

-01

May

-01

Jun-

01Ju

l-01

Aug

-01

Sep

-01

Oct

-01

Nov

-01

Dec

-01

Jan-

02Fe

b-02

Mar

-02

Apr

-02

May

-02

Jun-

02Ju

l-02

Aug

-02

Sep

-02

Oct

-02

Nov

-02

Dec

-02

Jan-

03Fe

b-03

Mar

-03

Apr

-03

May

-03

Jun-

03Ju

l-03

Aug

-03

Sep-

03O

ct-0

3N

ov-0

3D

ec-0

3Ja

n-04

Feb-

04M

ar-0

4A

pr-0

4M

ay-0

4Ju

n-04

Jul-0

4A

ug-0

4S

ep-0

4O

ct-0

4N

ov-0

4D

ec-0

4Ja

n-05

Feb-

05M

ar-0

5A

pr-0

5M

ay-0

5Ju

n-05

Jul-0

5A

ug-0

5S

ep-0

5O

ct-0

5N

ov-0

5

inde

x

GP'sSpecialistsOther service providersHospitalsMedicineTotals

MPI Hospital sub-categories

50

70

90

110

130

150

170

190

Jan-

01Fe

b-01

Mar

-01

Apr

-01

May

-01

Jun-

01Ju

l-01

Aug

-01

Sep

-01

Oct

-01

Nov

-01

Dec

-01

Jan-

02Fe

b-02

Mar

-02

Apr

-02

May

-02

Jun-

02Ju

l-02

Aug-

02S

ep-0

2O

ct-0

2N

ov-0

2D

ec-0

2Ja

n-03

Feb-

03M

ar-0

3Ap

r-03

May

-03

Jun-

03Ju

l-03

Aug

-03

Sep-

03O

ct-0

3N

ov-0

3D

ec-0

3Ja

n-04

Feb-

04M

ar-0

4A

pr-0

4M

ay-0

4Ju

n-04

Jul-0

4A

ug-0

4S

ep-0

4O

ct-0

4N

ov-0

4D

ec-0

4Ja

n-05

Feb-

05M

ar-0

5A

pr-0

5M

ay-0

5Ju

n-05

Jul-0

5A

ug-0

5S

ep-0

5O

ct-0

5N

ov-0

5

Hospital - non surgical consumables

Hospital - ward fees

Hospital - surgical consumables

Hospital - theatre fees

MPI Medicine category and sub-categories

80

85

90

95

100

105

110

115

120

125

130

Jan-

01Fe

b-01

Mar

-01

Apr

-01

May

-01

Jun-

01Ju

l-01

Aug-

01Se

p-01

Oct

-01

Nov

-01

Dec

-01

Jan-

02Fe

b-02

Mar

-02

Apr

-02

May

-02

Jun-

02Ju

l-02

Aug

-02

Sep-

02O

ct-0

2N

ov-0

2D

ec-0

2Ja

n-03

Feb-

03M

ar-0

3A

pr-0

3M

ay-0

3Ju

n-03

Jul-0

3Au

g-03

Sep-

03O

ct-0

3N

ov-0

3D

ec-0

3Ja

n-04

Feb-

04M

ar-0

4A

pr-0

4M

ay-0

4Ju

n-04

Jul-0

4A

ug-0

4Se

p-04

Oct

-04

Nov

-04

Dec

-04

Jan-

05Fe

b-05

Mar

-05

Apr

-05

May

-05

Jun-

05Ju

l-05

Aug-

05Se

p-05

Oct

-05

Nov

-05

Medicine - ChronMedicine - AcuteMedicine

Limitations• Submitted medical scheme claims – not total

spend• Specific medical schemes• Coding practices• Substitution effects – frequency of weighting

changes• Other influences:

– Benefit adjustments– Demographic changes– Regulatory changes– Medical practice– Provider agreements

MCI Adjustments

• Utilisation trends are driven by a number of factors including:– Benefit design;– Changing demographics;– Changing treatment patterns and practices;– Attitudes toward claiming;– Member plan movements;– Changes in medical technology.

MUI

• Calculated from MPI and MCI• MUI = MCI / MPI• Assumes that change not attributable to

price are utilisation• MCI is after adjustment

MCI/MPI = MUI Results

80

90

100

110

120

130

140

150

Jan-

01Fe

b-01

Mar

-01

Apr

-01

May

-01

Jun-

01Ju

l-01

Aug

-01

Sep-

01O

ct-0

1N

ov-0

1D

ec-0

1Ja

n-02

Feb-

02M

ar-0

2A

pr-0

2M

ay-0

2Ju

n-02

Jul-0

2A

ug-0

2Se

p-02

Oct

-02

Nov

-02

Dec

-02

Jan-

03Fe

b-03

Mar

-03

Apr

-03

May

-03

Jun-

03Ju

l-03

Aug

-03

Sep-

03O

ct-0

3N

ov-0

3D

ec-0

3Ja

n-04

Feb-

04M

ar-0

4A

pr-0

4M

ay-0

4Ju

n-04

Jul-0

4A

ug-0

4Se

p-04

Oct

-04

Nov

-04

Dec

-04

Jan-

05Fe

b-05

Mar

-05

MCIMPIMUI

Further work….

• More extensive data pool– Coding consistency!

• Impact of alternative reimbursement mechanisms

• Tracking utilisation• Measuring quality

INDEXATION OF MEDICAL COSTS FOR SOUTH AFRICAN MEDICAL SCHEMES

Roseanne da Silva

IAAHS 2007IAA Health Section Colloquium

13th – 16th May 2007CTICC

www.iaahs2007.com