The role of patient-physician communication on healthcare costs

Purpose and Background.  Healthy People 2020 provide comprehensive set of national goals and objectives for improving the health of all Americans. Primary care providers (PCPs) play an important role in protecting the health and safety of the communities they serve. Having a usual PCP is associated with greater patient’s trust in the providers, better patient-provider communication and increased likelihood that patients will receive appropriate care. This paper empirically investigates the role of patient-physician communication on the likelihood of receiving appropriate care, and its effect on healthcare costs at hospital inpatient settings.

 Data and Methods. The Healthcare Cost and Utilization Project (HCUP) 2008-2010 State Inpatient Databases (SID) for Florida (FL) were used in this analysis. The SID contain detailed diagnoses and procedures, total charges and patient demographics including gender, age, race and expected payment source. The SID also provide unique patient and physician identifiers. To obtain costs, we applied hospital specific HCUP cost-to-charge ratios to charges. We adjusted these costs with the CMS area wage index. Then, we linked Florida SID with Physician Licensure Database (PLD) obtained from Florida Medical Board of Examiners using physicians’ licensure numbers. PLD provide unique information on physicians, i.e. medical schools, graduation date, field of practice, languages. We obtained information about hospital characteristics using the American Hospital Association Annual Survey Database; and county level information from the Area Resource File. Our key covariate of interest is the association between Spanish speaking physicians and Hispanic patients with the total costs associated with that visit. We created a dummy variable -with value one if patient is Hispanic and physician in charge can speak Spanish; with value zero if patient is Hispanic and physician in charge cannot speak Spanish. Our methodological approach controls for patients heterogeneity, and reduces biases resulting from aggregation over patients over time. To assess the robustness of our baseline results, we conducted several empirical estimations and tested their significance. We started with descriptive analysis. Next, we used logistic regression models to assess likelihood of choosing a Spanish speaking physician over other physicians. Next, we used generalized linear regression models (GLM) to estimate and then compare the healthcare costs associated Hispanic patients with Spanish speaking physicians against others. We also used Oaxaca-Blinder (OB) Decomposition to compare the healthcare costs between and within Hispanic and non-Hispanic white patients across Spanish and non-Spanish speaking physicians.

 Results. We found a strong correlation at hospital inpatient settings between Hispanic patients and Spanish speaking physicians. Our risk adjusted estimates show that the odds ratios for Hispanic patients registered to Spanish speaking physicians is 3.8 compared to non-Hispanic white patients registered to Spanish speaking physicians. We found that hospital inpatient costs associated with Hispanic patients registered to Spanish speaking physicians is about $650 less relative to Hispanic patients registered to non-Spanish speaking physicians. Finally, our risk-adjusted results show that hospital inpatient costs associated with non-Hispanic white patients registered to non-Spanish speaking physicians relative to the non-Hispanic white patients registered to Spanish speaking physicians are lower by about $700 per visit.