Project Abstract

Abstract
Health care risk adjustment is a critical component of the healthcare system that aims to fairly compensate healthcare providers based on the health status of their patients. Traditionally, risk adjustment models have relied on demographic and diagnostic information to predict healthcare costs and outcomes. However, there is increasing recognition that non-traditional variables, such as social determinants of health, may play a significant role in predicting healthcare utilization and costs. This research project explores the impact of non-traditional variables in health care risk adjustment. By analyzing data from a large health insurance claims database, we aim to evaluate the effectiveness of incorporating non-traditional variables, such as housing instability, food insecurity, and transportation barriers, into existing risk adjustment models. We hypothesize that these non-traditional variables can improve the accuracy of risk adjustment by capturing additional dimensions of patient health and social needs that are not captured by traditional variables alone. Our study employs advanced statistical techniques, including machine learning algorithms, to develop and validate risk adjustment models that incorporate non-traditional variables. We will compare the performance of these models against traditional risk adjustment models in terms of their ability to predict healthcare costs and outcomes. In addition, we will conduct subgroup analyses to assess whether the impact of non-traditional variables varies across different patient populations. The findings of this research have the potential to inform policy decisions regarding the inclusion of non-traditional variables in health care risk adjustment. By demonstrating the value of incorporating social determinants of health into risk adjustment models, our study may pave the way for more equitable reimbursement practices that account for the diverse needs of patient populations. Furthermore, our research may help healthcare providers identify high-risk patients who may benefit from targeted interventions aimed at addressing social determinants of health. Overall, this project contributes to the growing body of literature on the role of non-traditional variables in health care risk adjustment. By highlighting the importance of considering social determinants of health in risk adjustment models, our research underscores the need for a more comprehensive and inclusive approach to assessing and compensating healthcare providers based on patient health status.

Project Overview

INTRODUCTION

1.1              Background of the study

The business of risk adjustment has come a long way since the publication of the Academy’s “Monograph Number One” with the title, “Health Risk Assessment and Health Risk Adjustment—Crucial Elements in Effective Health Care Reform” in May 1993. Less than ten years later, we had hospital inpatient diagnosis-based approaches, such as the model used by the Market Stabilization Pool for small group and individual coverage in NYS in conjunction with mandated community rating. The PIP-DCG approach for Medicare + Choice, also inpatient only, soon followed.

Risk adjustment models have included variables such as demographic (i.e. age and gender) and clinical markers based either on ICD-9 diagnosis codes and/or pharmacy codes such as the National Drug Codes (NDCs). Literature points to other variables such as geography, Body Mass Index (BMI), education, and income that also explain the variation in healthcare cost – but have hitherto not been included in risk adjustment programs mainly because such variables are not typically found in claim data. If these nontraditional variables explain meaningful variation in cost beyond traditional risk adjustment models – then this may provide incentives for issuers to select certain members. If such incentives lead to selection that affects the financial performance of issuers – then the policy goals of the risk adjustment program will be undermined. Recognizing the importance of fortifying risk adjustment programs against selection based on nontraditional variables, the Society of Actuaries’ Health Section sponsored an in-depth study into the relationship of nontraditional variables with health costs. This report presents the results of this study. We used the Medical Expenditure Panel Survey (MEPS) data in this research. Specific details concerning the data and preparation can be found in Section 3.4. This data is unique in that it includes a large number of individual characteristics (from BMI to whether a person has difficulty enjoying hobbies) together with healthcare claim data. There are limitations to the use of MEPS data, and these limitations are discussed further in Section 4. The results of this research demonstrate that it is important to adjust the traditional risk adjustment model in order to recognize nontraditional variables. The report develops a new measure (Loss Ratio Advantage or LRA) to help quantify the potential of a nontraditional variable to affect a risk adjustment program. With the help of this measure, the report compares the importance of over thirty variables that were systematically narrowed down from a list of over fifteen hundred variables describing various characteristics of the general population (i.e. the purchasers of healthcare insurance coverage). The nontraditional variables were broadly categorized into (1) demographic, (2) economic, (3) lifestyle, (4) psychological self-assessment (i.e. how a person feels about their mental health), and (5) physical self-assessment.

1.2              Statement of the problem

Risk adjustment of any kind is inherently imperfect, the complexity and sophistication of risk adjustment models has increased significantly in the past couple decades. With the passage of the Affordable Care Act (ACA), risk adjustment will be required for non-grandfathered commercial small group and individual coverage both inside and outside Exchanges. Using a structured and scientific approach, the researcher has examined a long list of non-traditional drivers of health cost, chosen the most relevant ones, and tested their effect on bottom-line medical cost when included in the traditional risk adjustment formula.

1.3       Objectives of the study

1. To determine the relationshipbetween non-traditional variables and health care risk adjustment in Nigeria.

4. To ascertain the impact of non-traditional variables on health care risk adjustment in Nigeria.

1.4       Research questions

1. Is there a relationship between non-traditional variables and health care risk adjustment in Nigeria?

4. Does non-traditional variables significantly impacts on health care risk adjustment in Nigeria?

1.5       Research hypotheses

Ho: There is no relationship between non-traditional variables and health care risk adjustment in Nigeria.

Hi: There is a relationship between non-traditional variables and health care risk adjustment in Nigeria

Ho: Non-traditional variables have no significant impact on health care risk adjustment in Nigeria.

Hi: Non-traditional variables significantly impacts on health care risk adjustment in Nigeria.

1.6       Significance of the study

The Affordable Care Act (ACA) includes the mechanism of risk adjustment in commercial small group and individual markets in order to further the policy goals of premium stabilization, mitigating incentives for issuers of healthcare coverage policies (issuers) to avoid unhealthy members, and to remove any advantages or disadvantages for plans inside healthcare exchanges compared to plans outside of such exchanges. The importance of risk adjustment to these policy goals cannot be overemphasized, and details such as the variables that are included in the risk assessment formula affect the extent to which the program is successful in meeting these goals.

1.7       Scope of the study

The study focuses on the impact of non-traditional variables in health care risk adjustment in Nigeria, University of Uyo Teaching Hospital (UTH) in Uyo Local Government Area of Akwa Ibom state was used as the case study.

1.8       Limitations of the study

This study has some limitations most especially in the area of data collection. Financial constraints as well as time available for the completion of the study are among other factors that would limit the scope of the study.

1.9       Definition of terms

Health Care: The organized provision of medical care to individuals or a community.

            Non-traditional: Not conforming to or in accord with tradition.

Risk Adjustment: A concept that refines an investment’s return by measuring how much risk is involved in producing that return.