IHP Final Project PDF

Preview

Summary

Final Project...

Description

Ashley Hicks Southern New Hampshire University IHP 340 Physician Communication Survey Analysis Review

PHYSICAN COMMUNICATION SURVEY ANALYSIS REVIEW

Page |2

This analysis review will be done on the research article, “The Effects of Hospital-Level Factors on Patients' Ratings of Physician Communication” published in the Journal of Healthcare Management. The purpose of the study is to look at the possible outside influences that could have a negative effect on scoring when it comes to physician communication surveys. Statistical software used in the regression analysis of the data was Stata 13 from StataCorp . The actual research question being asked within the article is how does the organizational-level factors influence physician communications with their patients in the inpatient setting. Many outside factors could come into play when it comes to physician communication with the patient that are well out of the physician’s control such as the actual organizational structure of the hospital, the staffing ratios for not only the physician but also for the nurses as well and all of these things have to be taken into account when looking into the scores given for physician communication. The hypothesis for this study is, “that a higher physician staffing level, a key structural attribute, is associated with higher ratings of physician-patient communication.” [ CITATION AlA16 \l 1033 ]. For the research methods used within this observational study there were two sources of information gathered and used within the research, the first is the information complied from the 2009 Hospital Consumer Assessment of Healthcare Providers and Systems, HCAHPS as it will be henceforth referred to, to as well as the American Hospital Associations annual, herein referred to as the AHA, survey database for the same year. I came to the conclusion this was an observational study as the study took the data from two different data sources to observe the variables obtained within them then determined if there was any correlation thusly making in an observation study rather than an experimental one as they did not try to control any of the

PHYSICAN COMMUNICATION SURVEY ANALYSIS REVIEW

Page |3

variables within the study at any point in time to find a causality within the results. The best example within the study would be the following; “For our analysis, we merged the AHA annual survey database (2009) with HCAHPS data. All organizational level variables were derived from the AHA database. Specifically, we retrieved information about hospital size, FTE physicians and dentists per 10,000 inpatient days, hospitalists providing care at the hospital, physician ownership, for-profit ownership, public ownership, teaching status, specialty status, Medicare share of admissions (i.e., Medicare admissions divided by total admissions), EHRs, and high technology index.” [ CITATION AlA16 \l 1033 ] This shows that the information was observed rather than conducted within an experiment. The methods that were used to collect the data are pretty simple considering the data was extracted from surveys that had already been completed for the 2009 year but both come from reputable information sources such as the Centers for Medicare & Medicaid Services Hospital Compare as well as the American Hospital Association websites respectively. However, the data that was used within the research article was the data that came from the bottom of the HCAHPS survey or within the last 25th percentile. This was the data that showed the “worst” of physician communication or was shown to be within the range that stated the physician “sometimes or never” communicated well with the patient as the study states this was a variable that was reported within the actual HCAHPS survey. The Centers for Medicare and Medicaid Services website states that; “The HCAHPS survey asks discharged patients 29 questions about their recent hospital stay. The survey contains 19 core questions about critical aspects of patients' hospital

PHYSICAN COMMUNICATION SURVEY ANALYSIS REVIEW

Page |4

experiences (communication with nurses and doctors, the responsiveness of hospital staff, the cleanliness and quietness of the hospital environment, communication about medicines, discharge information, overall rating of hospital, and would they recommend the hospital). The survey also includes three items to direct patients to relevant questions, five items to adjust for the mix of patients across hospitals, and two items that support Congressionally-mandated reports.”. [ CITATION HCA20 \l 1033 ] So, of these 29 questions roughly 19 of the contain information about the communication of the actual physician so just above 65% of the information. As for the AHA survey the AHA data website states the following about the information contained therein: “The AHA Annual Survey of Hospitals, profiles a universe of nearly 6,400 hospitals throughout the United States and associated areas, with exclusive data on health system performance and measures on patient satisfaction, readmissions, mortality, and quality.”[ CITATION Dat \l 1033 ] Given the research question of, “how does the organizational-level factors influence physician communications with their patients in the inpatient setting.”[ CITATION AlA16 \l 1033 ]. The HACHP survey along with the AHA data was the appropriate way to go about finding the information needed for the study. The information used within the study is categorical given that the questions asked of the original participants were asked to rate the communication from their physician withing the following scale; “Also, our outcome variable-patients' ratings of physician communication-is measured in HCAHPS by means of a frequency scale ranging from "sometimes or never" to "always,"

PHYSICAN COMMUNICATION SURVEY ANALYSIS REVIEW

Page |5

rather than by asking patients to rate the quality of communication using scale points such as "poor" or "excellent." This can limit the validity of the measure because "always" is not necessarily the same as "excellent".” [ CITATION AlA16 \l 1033 ] This would mean could possibly use a histogram to show the data results with the X axis showing the different columns of answers, such as “poor communication” to “excellent communication” and thusly the Y axis showing the actual number of participants to who answered within the respective column. There are many potential weaknesses for the data in that is that of the questions asked on the HCAHPS only 19 pertain to physician communication but the other 35% or so that are unrelated could vary well bring the overall score down if a patient was unhappy/disgruntled about something unrelated. Another potential weakness would the physician staffing level variable that the study states is, “similar to the commonly used measure for nurse staffing, this measure, to our knowledge, has not been tested or validated in other studies”[ CITATION AlA16 \l 1033 ]. As for the key demographics of the population it is simply people to whom have recently discharged from the hospital with no real qualifiers on who got the survey just everyone who was recently discharged to whom had been inpatient. The key findings within the research state that there are three main hospital level factors that “have a statistically significant association with patient ratings of physician communication at a .05 significance level”[ CITATION AlA16 \l 1033 ]. Those three factors are the actual size of the hospital, if there were hospitalists providing the patient care, and the for-profit ownership of the hospital that factored negatively to the scores. Of those three factors the most significant is seemingly if there were hospitalists providing the care for the patient, which is directly related to physician staffing levels. Physician staffing levels is as the study states a significant factor that

PHYSICAN COMMUNICATION SURVEY ANALYSIS REVIEW

Page |6

has yet to be explored but given the fact that physician burn out is higher now than ever, (Shanafelt et al., 2012), so this puts even more emphasis on the results of the study considering there is an almost direct correlation between patient laden physician staffing models and lower physician communication scores. Hospitals that were for-profit as well as larger hospitals also had lower physician communication scores reported, which could be seen as a direct correlation to physician staffing models as well. The results within the research did turn out to be statistically significant as they were able to use the quantitate data given to show which variables outside the physician’s control directly contributed to the lowered physician communication scores....