Harmonizing Healthcare Financing for Health Equity: Case Studies of Cross-subsidization in Thai Public Hospitals 2015.6.6

Harmonizing Healthcare Financing for Health Equity: Case Studies of Cross-subsidization in Thai Public Hospitals

Asst. Prof. Borwornsom Leerapan, MD PhD

JCMS2015: Health Equity Through Innovation and Collaboration June 6th, 2015

Pix source: Jirapat Mobkhuntod “Treatment of human equality” (PMAC 2015 World Art Contest, 9-13 years old)

Special thanks to:

Ø Phatta Kirdruang, PhD

Ø Thaworn Sakulpanich, MD, MSc.

Ø Patchanee Thamwanna

Ø Utoomporn Wongsin

Ø Nutnitima Changprajuck

Ø Health Insurance System Research Office (HISRO) & Health System Research Institute (HSRI)

①  Concept of cross-subsidization

–  Who cross-subsidies what to whom?

②  Case studies of Cross-subsidization in Thai Public Hospitals

–  Study objectives, methods, results

③  Implications for policy and practice

–  What’s next for researchers & policymakers?

Presentation Outline

Pix source: online.wsj.com 

Financing of Thai Healthcare Systems CSMBS SSS UCS Motor Vehicle

Victim Protection Law

Private Health Insurance

Feature State/Employer welfare

Compulsory heath insurance with state subsidies

State welfare Compulsory heath insurance for vehicle owners

Voluntary health insurance

Targeted groups of beneficiaries

Civil servants, state enterprise employees and dependents

Employees in private sector and temporary employees in public sector

Thai citizens without the coverage of CSMBS & SSS

Victims of vehicle accidents

General public

Source of financing

Govt. budget

Tri-party (Employee, employer and govt. budget)

Govt. budget

Vehicle owners Household

Method of payment to health facilities

Fee-for-service Capitation and Fee-for-service

Capitation and Fee-for-service

Fee-for-service Fee-for-service

Major problems Rapidly and constantly rising costs

Covering while being employed only

Inadequate budget

Redundant eligibility and slow disbursement

Redundant eligibility and slow disbursement

Source: Adapted from Wibulpolprasert et al. (2011). Thailand Health Profile 2008-2010.

CGD (CSMBS),

NHSO (UCS)

Taxes Payers

Employer-based private health

insurance

Individual & Employer’s

private health insurance

(Voluntary)

Hospitals

Medical Specialists

Generalists & PCPs

Patients paying out-of-pocket

Ambulatory Facilities

Payment Mechanisms: Salary, Fee-for-Service,

Global Budget, Capitation, DRGs, etc.

Financing of Thai Healthcare Systems

Providers in Public & Private Sector

Commercial Insurance

Companies

Social Security

Office (SSS)

Motor vehicle’s owners (Mandatory by the Motor

Vehicle Victim Protection Law)

Study Rationale: High Expenditures?

Figure  source:  Benjaporn  (2007)  14

of the former period grew about 68 percent, 26 percent and 41 percent, respectively,

while during the latter period they were about 224 percent, 43 percent and 127 percent.

Furthermore, when consider growth of expenditure compare year on year; growth rate

of the out-patient expenditure during the second period showed an upward trend and

had very rapid growth in the last two years, 2006 and 2007 (graph 2.5).

With respect to expenditure per patient, this study can merely consider the average in-

patient expenditure, because of data limitations. According to data from the electronic

payment system, the average in-patient expenditure in 2003-2006 increased over time as

shown in graph 2.6.

Graph 2.4: CSMBS expenditure during the fiscal years 1996-2007

Source: The Comptroller General’s Department and the Government Fiscal

Management Information System (GFMIS)

Note: 1 Euro = 49.4450 Baht, as of January 8, 2008

4,826 5,625 5,866 6,206 7,007 8,123

9,509

11,350 13,905

16,943

21,896

30,833

8,761 9,877 10,574 9,048 10,050 11,058 10,967

11,335 12,138 12,437 15,109

15,649

13,587 15,502

16,440 15,253

17,058 19,181 20,476 22,686

26,043

29,380

37,004

46,481

0 5,000

10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 50,000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Mill

ion

Bah

t

Year

CSMBS Expenditure in the fiscal years 1996-2007

Out-patient In-patient Total

Ø  Common assumptions of what causes increasing healthcare expenditures of CSMBS: •  Overuse of NED drug?

•  Overuse of brand-named drugs?

•  Limited EBM practices?

•  Corruption in healthcare sector?

Ø Cross-­‐subsidiza,on  can  be  a  missing  piece!    

Study Rationale: Cross-subsidy?

Figure  source:  www.be2hand.com;  www.imdb.com;  workwithbrianandfelicia.com  

Ø  “Do hospitals use payments of a type of health services to subsidize or support financing of other services?”

•  If so, how?, at which level?, at what degree?

Literature Review

Ø  Concepts of “cross-subsidization” or “cost-shifting” from developed countries such as the U.S. (Morrisey 1994, Cutler 1998, Dranove 1988, Feldman et al. 1998, Frakt 2010 & 2011).

Ø  Such theorectical concepts might not be applicable in Thailand’s healthcare systems, especially that Thai public hospitals do not have the “ability to set prices” by themselves.

Ø  There was no empirical study of cross-subsidization in the contexts of Thai healthcare systems.

Study Objectives

1.  To explore motivations and existing practices of the administrators of Thai public hospitals that potentially can lead to cross-subsidization (“to use payments of a type of health services to support financing of other services”).

2.  To demonstrate an empirical evidence related to cross-subsidization at the hospital level, including the cost difference and the difference of excess of revenues over expenses among health schemes.

Methodology: Research Design

Ø No existing empirical studies of cross-subsidization in the contexts of Thai healthcare system.

Ø  Theoretical concepts developed in countries like the U.S. might not be applicable in Thailand.

Ø Mixed-­‐methods  research  (MMR)  with  the  concurrent  embedded  research  design  (Creswell  et  al.,  2009).    

Source: Creswell (2009). Research design: Qualitative, quantitative, and mixed methods approaches. 3rrd ed.

Ø  Primarily,  the  QUAL  analysis.  

Ø  The  QUAN  analysis  is  used  to  compliment  the  QUAL    analysis.  

Methodology: Source of Data

Ø Data was based on three selected public hospitals: Ø  Two medical centers with 1,000 and 1,134 beds

Ø One teaching hospital with 1,378 beds.

Ø Hospitals were purposefully selected, based on the accessibility to the hospital administrators and the availability of the datasets of unit cost, claims, and reimbursement.

Methodology: Data Collection

Ø Qualitative data: Ø  Semi-structure interviews and focus-group interviews.

Ø  30 key informants who are responsible for the administration of the three hospitals.

Ø  Verbatim was transcribed and analyzed using ATLAS.ti 7.

Ø Quantitative data: Ø  Secondary data of inpatient care, collected at the patient level,

from the two medical centers.

Ø  Unit-cost, charge, reimbursement, patient’s health scheme, DRG codes, and basic demographic characteristics.

Ø  Analysis was conducted using Stata 12.

Qualitative Analysis

Ø Constructivist grounded theory (Chamaz, 2005; 2006) Ø Coding process (Strauss & Corbin 1990)

Ø Analyze the cost differences across health schemes Ø By using descriptive statistics and a regression analysis.

Ø Compare the differences among charge, cost, reimbursement, particularly ‘reimbursement-cost’ and ‘reimbursement-to-cost ratio’: Ø Across health schemes

Ø Across MDC groups

Ø Across Age groups

Quantitative Analysis

Ø 13 sub-themes, categorized into 4 emerging themes.

Qualitative Findings

Sub-themes Themes

Varied understanding of cross-subsidization, Unclear financing for non-healthcare missions

No mutual understanding or attitudes toward concepts and practices of cross-subsidization

Inadequate reimbursement, Non-performing loan, Unequal negotiation power

Obstacles facing management due to policies of the payers

Conflicting roles between quality & equity-focus and efficiency-focus, Limited information to manage prices and cost

Obstacles facing management due to organizational limitations

To be missions-driven organization, To focus more on efficiency than revenues, To do public funds raising, To control the volume of certain groups of patients when feasible, To advocate changes of the payer’s policies

Organizational responses to policies of the payers

#1 No Mutual Understanding or Attitudes

#1 No Mutual Understanding or Attitudes

#2 Obstacles of Financial Management

#2 Obstacles of Financial Management

#3 Responses Directly to Payer’s Policy

Quantitative Findings #1: Cost Differences across Health Schemes

“Total Cost Across Health Schemes”

010

,000

20,0

0030

,000

mea

n of

tota

lcos

t

CSMBS SSS UC Cash

Source:  Center  hospital  #1  

Ø   The average costs per visit vary across health schemes, where CSMBS patients have the highest cost.

Ø  After controlling for age, gender, disease, LOS, the regression analysis confirms that the patient’s health scheme has a significant impact on the unit cost of health services.

Quantitative Findings #2: Profit or Loss across Health Schemes

“Difference between Reimbursement and Cost” (by Health Scheme)

-10,

000

-5,0

000

5,0

00

me

an o

f rei

mb_

cost

_diff

csmbs sss uc foreign cash Others

Source:  Center  hospital  #2  

Ø   Assume that charge equals reimbursement for foreign, OOP, and ‘others’ groups.

Ø  Reimbursement (or charge) is much lower than the cost for UC and foreign patients. Ø  Insufficient reimbursement Ø  Hospital’s burden to take care of patients without health insurance or rights (e.g. foreign patients)

“Difference between Reimbursement and Cost” (by DRG-MDC)

-30,00

0-20

,000

-10,00

00

10,000

mean

of reim

b_cost

_diff

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 28

Source:  Center  hospital  #1  

Ø   The hospital receives reimbursement more than the cost for only 5 MDC groups. Ø  Some major diagnostic categories create a large deficit for the hospital.

MDC  5  =  Diseases  &  disorders  of  the  circulatory  system  

MDC  22  =  Burns  

Quantitative Findings #2: Profit or Loss across Health Schemes

Quantitative Findings #2: Profit or Loss across Health Schemes

“Difference between Reimbursement and Cost” (by Health Scheme and Age group)

-5,0

000

5,00

010

,000

<20 21-30 31-40 41-50 51-60 61-70 71+

mean of reimb_cost_diff_CS mean of reimb_cost_diff_SSmean of reimb_cost_diff_UC mean of reimb_cost_diff_cash

Source:  Center  hospital  #1  

Ø   ‘Reimbursement-Cost’ is generally positive for CSMBS, and the difference is large for elder patients.

Ø  This difference is negatives for almost all age groups for UC patients.

•  Given our limited datasets, we found no direct evidence that suggests hospitals cost-shift by increasing prices charged to out-of-pocket payment (OPP) patients to compensate for the loss.

•  Yet, patterns of decision-making of hospital administrators related to cross-subsidization were found.

•  Therefore, financing policies of one health scheme also impact other patients groups within the hospitals.

What Do Findings Tell Us?

Mental Models of Hospital Administrators

Implications for Policy and Practice

Ø  To policymakers: •  Demonstrates an empirical evidence of

that current healthcare financing of hospitals still inappropriate/inadequate.

•  Suggests that payments from particular payers potentially can be used as a “buffer” for hospitals, potentially leading to “passive cross-subsidization” and inequity issues of healthcare access.

•  Suggests the needs to “harmonize” healthcare financing in a more efficient and equitable fashion.

Q & A [email protected]

Pix source: online.wsj.com