ASSESSMENT OF THE AMBULATORY CARE WORKFORCE IN GREATER NEW ORLEANS

Transcription

1 ASSESSMENT OF THE AMBULATORY CARE WORKFORCE IN GREATER NEW ORLEANS RESULTS FROM A SUMMER 2007 SURVEY OF HEALTH CARE PRACTICES IN THE GREATER NEW ORLEANS AREA APRIL 2008 Bringing People, Ideas and Resources Together

2 ASSESSMENT OF THE AMBULATORY CARE WORKFORCE IN GREATER NEW ORLEANS RESULTS FROM A SUMMER 2007 SURVEY OF HEALTH CARE PRACTICES IN THE GREATER NEW ORLEANS AREA APRIL 2008 BY GREGORY STONE CLAYTON WILLIAMS CLAUDIA CAMPBELL MARK DIANA KARA WAYMAN Financial support provided by: Baptist Community Ministries HCA, Inc. The Fry Foundation

3 TABLE OF CONTENTS EXECUTIVE SUMMARY... 1 INTRODUCTION... 3 SURVEY DESIGN... 3 RESULTS... 5 STUDY LIMITATIONS DISCUSSION AUTHORS AND ACKNOWLEDGEMENTS ABOUT LOUISIANA PUBLIC HEALTH INSTITUTE REFERENCES APPENDIX A: SURVEY QUESTIONNAIRE APPENDIX B: MAPS OF SERVICE DELIVERY SITE LOCATIONS... 32

4 EXECUTIVE SUMMARY Hurricane Katrina (August 29, 2005) and the subsequent flooding caused by levee failures devastated the Greater New Orleans area. Residential and commercial property was severely damaged in the four parish region of Orleans, Jefferson, Plaquemines, and St Bernard parishes. 1 Evidence suggests that the storm s destruction displaced many physicians, damaged their practices, and left considerable unmet medical and mental health care needs among area residents one year after the storm. 2-5 Increasing numbers of displaced and new residents continue to repopulate the city. 6 They are likely to place additional pressures on the region s fragile health care system, which was severely crippled by the storm. 7 As recovery continues in greater New Orleans, it is critical to understand the capacity of the health care workforce to meet the needs of the growing population. The purpose of this report is to describe the capacity and characteristics of the office-based health care workforce that exists in the region and to consider how accessible it is to the population. The data for this report were collected through a survey conducted in the summer of 2007 of all operating clinics and office-based, health care delivery sites throughout the four parish region. Hospital outpatient and emergency departments and dental practices were not included. To identify sites to survey, a potential universe was compiled using public and private directories and by canvassing the neighborhoods. Questionnaires were mailed to every potential delivery site and hand delivered to every delivery site that was identified through canvassing. In-person visits were made to every potential site to determine whether it was currently operating and therefore eligible for the survey. This process identified 1238 eligible delivery sites. Eligible delivery sites were categorized as primary care, medical specialty, behavioral health, or other, according to the types of health care service provided. After questionnaires were mailed or hand delivered to every potential delivery site, a minimum of four follow-up visits were made to those sites that did not return a completed questionnaire by mail. The final survey response rate was 73.6 percent (911/1238). However, not every item was completed for every survey so the number of respondents varies by survey question. Key Findings: The majority of delivery sites are located in Jefferson Parish (65.1%). Only one-third (33.3%) are located in Orleans Parish and less than two percent (1.6%) are located in Plaquemines and St Bernard Parishes. Among responding sites, 70.7 percent are physician-based, meaning at least one physician worked at the site part or full time. The vast majority of primary care and medical specialty sites are physician-based (96.8% and 96.3%, respectively) while less than one-third (31%) of behavioral health sites include a physician on staff. The survey identified 1394 physician full time equivalents (FTEs) practicing in the greater New Orleans region in the summer of 2007, with physician FTEs per site ranging 1

5 from 0.1 to 40. An additional non-physician medical and behavioral health professional FTEs were found to be working at the responding sites. Of the 1394 physician FTEs identified by the survey, 39.2 percent (546.8) are primary care specialists. A nearly equal percentage, 38.8 percent (540.2), are medical specialists and 22 percent (307.1/1394) of physician FTEs are surgical specialists. Among physician-based responding sites, the average number of annualized patient encounters per physician FTE is Among behavioral health sites, the average annualized patient encounters per provider FTE is Over one-third (38.0%) of 787 survey respondents plan to hire additional medical staff within the next year, with the highest proportion occurring among primary care sites (47.8%) and sites in St Bernard Parish (85.7%). In Orleans Parish, nearly half (41.9%) of all responding sites plan to hire. Over 14 percent (111) of 779 survey respondents currently use electronic medical records (EMR). Other and primary care sites have the highest proportion using EMR (18.2% and 15.2%, respectively). Among the 769 respondents that were operating before Hurricane Katrina, 46.1 percent have experienced a decrease in patient volume from pre-storm levels. The decrease was most notable in St Bernard and Orleans parishes (85.7% and 59.8%, respectively) where considerable population displacement occurred due to storm damage and flooding. An estimate of 4.2 million patient encounters occurred at 724 responding delivery sites during a one-year period. More than two-thirds (68.9%) of encounters occurred in Jefferson Parish while almost one-third (29.6%) occurred in Orleans Parish and the remaining 1.5 percent occurred in Plaquemines and St Bernard parishes. Overall patients with private insurance accounted for 41.2 percent of encounters; Medicare, Medicaid, and uninsured patients accounted for 26.3, 15.1, and 17.4 percent, respectively. 2

6 INTRODUCTION Hurricane Katrina (August 29, 2005) severely devastated the Greater New Orleans area. Wind and water damage from the storm and the ensuing failure of levees that left 80 percent of the city underwater for several weeks caused severe damage to residential and commercial property in the four parish region of Orleans, Jefferson, Plaquemines, and St Bernard parishes. 1 The extent and severity of the destruction produced a mass migration which dramatically altered the region s demographics. 8 Evidence suggests that the storm s destruction displaced many physicians, damaged their practices, and left considerable unmet medical and mental health care needs among area residents over one year after the storm. 2-5 Elevated health care needs among an increasing population have placed additional pressures on the region s fragile health care system, which was severely crippled by the storm. 7 To inform sound recovery planning it is critical to understand the capacity of the health care workforce to meet the needs of the growing population. Early estimates suggest that some 4,500 active patient care physicians were displaced by the storm 2 and that displaced physicians face substantial financial obstacles to return. 3 More recent information indicates that the physician population has returned in proportion to the overall population in the region. 9 A gap exists in our knowledge of the number and types of active physicians, other health care professionals, and delivery sites in the region, whether they are operating at full capacity, and how accessible they are to the population. In response to this information gap, the Louisiana Public Health Institute conducted a survey of medical and mental health care practices in the four parishes comprising the greater New Orleans area between April and August of The survey was conducted in collaboration with the Louisiana Department of Health and Hospitals and Tulane University Health Sciences Center with financial assistance from Baptist Community Ministries, HCA Inc., and the Fry Foundation. The purpose of this report is to describe the capacity and characteristics of the office-based health care workforce that exists in the region and to consider how accessible it is to the population. SURVEY DESIGN The unit of analysis for the survey was a service delivery site, which was defined as an organization that includes one or more licensed medical or behavioral health practitioner that currently provides direct patient care at a specific facility or office space. Hospital emergency and outpatient departments and dental practices were not included in the survey. The delivery site was used as the survey population because it addresses two common issues associated with office-based practices: (1) a practice can have more than one physical location and (2) more than one practice can operate at the same physical location. The survey population was composed of potential delivery sites that were compiled from several public and private directories, including Medicaid providers, the Yellow Pages, the local Sunshine Pages directory, the VIA LINK resource directory, and a private mailing list of 3

7 health care facilities. Neighborhood canvassing and enumeration techniques were used to identify additional delivery sites that had not been included in the survey population. Facilities listed in the population that did not meet the eligibility definition of a service delivery site (i.e., administrative offices that do not provide patient care, vendors of health care supplies, and social services organizations that do not provide medical or behavioral health services) and delivery sites that were not operating during the survey period were excluded from the survey population. Questionnaires were mailed to every potential delivery site in the population and hand delivered to every delivery site identified through canvassing. The questionnaire consisted of 21 questions addressing the following areas: clinical services, patient volume and payer mix, use and barriers to use of electronic medical records, staffing, and the effects of Hurricane Katrina. A copy of the survey questionnaire is provided as Appendix A. Completed questionnaires were obtained by mail and in person. A minimum of four follow-up visits were made to obtain a completed questionnaire. Telephone follow-up was conducted in the case of incomplete or inconsistent responses. Once the survey process was complete, responding and non-responding delivery sites were classified according to the predominant type of health service provided: primary care, medical specialty care, behavioral health, and other. For respondents, the predominant types of physicians and health professional FTEs in the practice were used to classify the delivery site. For the non-respondents, phone book, internet sources, and information collected from survey visits were used to classify the type of care. Maps of the delivery site locations by service categories are provided as Appendix B. Chart 1 provides a breakdown of the number and types of delivery sites in each service category. The 389 sites that were categorized as other include chiropractic, eye care, laboratory, specialty care such as assisted living facilities and nursing homes and other categories of post acute care, such as wellness centers, fertility clinics, sleep clinics, rehabilitation clinics, and home health. Chart 1: Number of service delivery sites by category 277 Grouped as Other Behavioral Health Medical Specialty Primary Care Other Specialty Care Eye Care Chiropractic Lab 4

8 Delivery sites were stratified by parish and service category. Bivariate analyses using Fisher exact test and likelihood ratio χ 2 were conducted to assess differences in survey response by parish and service category, respectively. All data analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC). RESULTS The survey was administered to 1238 eligible delivery sites identified from the potential universe of sites in the area. Surveys were obtained from 911 (73.7%) delivery sites. Table 1 provides a breakdown of the response rates among all delivery sites by parish and service category. Table 1: Characteristics of survey respondent and non-respondent office-based delivery sites Category Respondents Non-respondents Total Number of Number of Number of Percent Percent Sites Sites Sites Percent Overall Parish Jefferson Orleans Plaquemines St Bernard Service category Primary Care Medical Specialty Behavioral Health Other Summary of Table 1 The response rates varied between 100 percent in St. Bernard Parish and 63.6 percent in Plaquemines Parish, the two parishes with the fewest number of sites. However, bivariate analysis revealed no statistically significant differences in survey response by parish (p=.11). Differences in response rates by service category were statistically significant (p=.005). Survey response was highest among primary care (79.4%) and medical specialty (77.1%) sites, while response was lowest among sites categorized as other (68.4%). After the minimum four survey visits were conducted, available resources allowed for an additional survey visit to all non- 5

9 responding primary care, medical specialty, and behavioral health sites, with an emphasis on obtaining staffing data from these sites at a minimum. These additional visits were not conducted for sites classified as other. The absence of this additional visit is a likely reason for a lower response rate among this category. Further, these visits resulted in some additional surveys with incomplete item responses, so the number of responses for each question listed in this report varies. Table 2: Characteristics of all office-based delivery sites surveyed in the region by location and category of service provided Category Number of Sites Percent a Parish Jefferson Orleans Plaquemines St Bernard Service category Primary Care Physician-based delivery site Non-physician-based delivery site Unknown if physician present Medical Specialty Physician-based delivery site Non-physician-based delivery site Unknown if physician present Behavioral Health Physician-based delivery site Non-physician-based delivery site Unknown if physician present Other Physician-based delivery site Non-physician-based delivery site Unknown if physician present a May not total 100 percent due to rounding 6

10 Summary of Table 2 Table 2 characterizes each delivery site by location and by presence or absence of a physician at the site. Because the presence of a physician FTE could not be verified for non-respondents, these delivery sites were classified as unknown. Of the 1238 delivery sites in the greater New Orleans region, the majority are located in Jefferson Parish (65.1%). Orleans Parish accounts for one third (33.3%) of the region s delivery sites while less than 2 percent are located in Plaquemines or St Bernard parishes. By service category, other represents the largest group (31.4%) followed by behavioral health (23.7%), medical specialty (22.5%), and primary care (22.4%). Table 3: Characteristics of physician and non-physician office-based delivery sites (911 respondents) Category Physician-Based Delivery Site Column Percent a Number of Sites Row Percent Non-Physician-Based Delivery Site Number of Column Row Sites Percent a Percent Overall Parish Jefferson Orleans Plaquemines St Bernard Service category Primary Care Medical Specialty Behavioral Health Other a May not total 100 percent due to rounding Summary of Table 3 Physician-based refers to whether or not a physician is on staff at the delivery site. Among the 911 sites that responded to the survey, the majority (70.7%) are physician-based. By service category, primary care and medical specialty sites are predominantly physician-based (96.8% and 96.3%, respectively) while less than one third (31.0%) of behavioral health sites are physician-based. Only Plaquemines Parish has a considerably smaller proportion (42.9%) of physician-based sites than the overall average. 7

11 Table 4: Full time equivalent (FTE) staffing at physician-based delivery sites (644 respondents) Category Average FTEs per Site Total FTEs Range of FTEs per Site by Type Physicians Primary Care Medical Specialty Behavioral Health Other Non-physician medical and behavioral health professionals Primary Care Medical Specialty Behavioral Health Other Administrative staff Primary Care Medical Specialty Behavioral Health Other Total staff Primary Care Medical Specialty Behavioral Health Other Summary of Table 4 Table 4 presents a breakdown of the workforce at physician-based delivery sites, which account for 70.5 percent of the survey responding sites (Table 3). At physician-based delivery sites, the survey identified 1394 physician FTEs and non-physician medical and behavioral health professional full time equivalents. The average number of 2.2 FTE physicians at physicianbased delivery sites is generally consistent across service categories, although behavioral health sites at 1.3 FTEs employ the fewest number of physician FTEs. In contrast, behavioral health and other sites that include physician staffing use more non-physician FTEs at 4.0 and 3.9, respectively, than the other two categories. Workforce size varies considerably by physicianbased sites as is evidenced by the wide range in minimum and maximum FTEs that staff each site. The average number of administrative staff FTEs varies only slightly by service category. 8

12 Administrative FTEs might include all non-provider staff such as clerks, billing managers, and office managers. Table 5: Full time equivalent (FTE) staffing at non-physician-based delivery sites (267 respondents) Average Minimum and Total Category FTEs per Maximum FTEs Site FTEs per site Non-physician medical and behavioral health professionals Primary Care Medical Specialty Behavioral Health Other Administrative staff Primary Care Medical Specialty Behavioral Health Other Total staff Primary Care Medical Specialty Behavioral Health Other Summary of Table 5 Non-physician-based delivery sites represent the minority of survey respondents (29.5%; Table 3). The survey identified 1553 medical and behavioral health provider full time equivalents at non-physician-based sites. The majority of these providers are located at sites categorized as other (80.1%). Among non-physician medical and behavioral health providers, full time equivalent averages vary considerably by service category, with other having the largest staffs (11.6) and behavioral health having the smallest (1.6). However, at least one extremely large behavioral health site was identified as evidenced by the wide range of FTEs. The absence of any non-physician medical or behavioral health professionals at one or more behavioral health and other delivery sites suggests that these sites are only staffed by physicians. The average administrative staff size of 0.6 FTEs at behavioral health sites indicates the presence of a relatively large number of small delivery sites in this service category. 9

13 Table 6: Full time equivalent (FTE) physicians at office-based delivery sites by specialty (911 respondents) Category Total FTEs from Respondents 95% Confidence Interval Estimated Total FTEs in Region a 95% Confidence Interval Primary care physicians ± ± 43.6 Family practitioners ± ± 17.5 General practitioners 60.3 ± ± 13.4 Internists ± ± 23.5 Pediatricians ± ± 22.9 Obstetricians/Gynecologists 82.0 ± ± 19.4 Other primary care specialists 63.5 ± ± 19.7 Medical specialist physicians ± ± 54.3 Cardiologists 34.2 ± ± 16.2 Dermatologists 45.7 ± ± 19.5 Neurologists 21.6 ± ± 8.7 Pediatric subspecialists 8.7 ± ± 4.9 Psychiatrists 90.7 ± ± 23.2 Other medical specialists ± ± 43.1 Surgeons ± ± 83.8 General surgeons 34.0 ± ± 11.7 Orthopedic surgeons 67.4 ± ± 19.6 Urologists 17.0 ± ± 8.4 Otolaryngologists 20.4 ± ± 8.7 Ophthalmologists 90.8 ± ± 63.5 Other surgical specialists 77.6 ± ± 33.8 Total physician FTEs ± ± a Total number of FTEs assuming that non-respondents have similar physician staffing characteristics as respondents Summary of Table 6 Primary care specialties and medical specialties make up and 540.2, respectively, of 1394 total FTE physicians and surgeons comprise FTEs. Of the primary care physicians, internists and pediatricians represent the largest specialty groups (131.6 and 108.9, respectively). Psychiatrists represent the largest group among medical specialists (90.7). Among surgeons, ophthalmologists are the largest group, followed by orthopedic surgeons (90.8 and 67.4, respectively). 10

14 The column headed Estimated Total FTEs in Region reports the number of office-based physician FTEs estimated to be practicing in the entire population in the four parish region. These estimates are obtained by weighting the survey data such that the survey respondent sites (73.6%) are assumed to be representative of the non-survey respondent sites (26.4%) in terms of physician staffing characteristics. Thus, the total number of physician FTEs identified in each service category in each parish are inflated to estimate the number of physician FTEs that would be found in the non-respondent sites in each corresponding service category and parish. The critical assumption made here is that the respondent and non-respondent sites in the same parish and service category are similar in physician staffing characteristics. Table 7: Average annual patient visits per full time equivalent (FTE) provider at office-based delivery sites (646 responses) Category Number of Sites Mean 95% Confidence Limit Annual patient visits per physician FTE ± Primary Care ± Medical Specialty ± Behavioral Health ± Other ± Annual patient visits per behavioral health provider FTE ± Primary Care ± Medical Specialty Behavioral Health ± 87.3 Other ± Annual patient visits per provider FTE ± 991 Primary Care ± Medical Specialty ± Behavioral Health ± 87.3 Other ± a a Insufficient sample size. Summary of Table 7 Table 7 presents the average annual patient visits per FTE provider in each type of delivery site. This ratio provides a measurement of workforce productivity. The highest level of productivity among physicians occurs in medical specialty delivery sites ( visits per FTE per year). The relatively lower productivity among physicians at 11

15 behavioral health sites (949.3) might be explained by the non-procedural nature of the patient encounter (e.g. psychiatric care and counseling), which generally requires longer visit times. In the table, behavioral health provider includes psychiatrists and other licensed mental health and substance abuse providers, which might include: psychologists, licensed clinical social workers, licensed marriage and family therapists, licensed addiction counselors, and licensed professional counselors. Behavioral health providers at primary care sites experience considerably higher productivity than in all other settings (5362.8). This might be explained by high need for brief counseling sessions and referral services in primary care settings and the nature of the patient encounters that occur at these sites. Table 8: Characteristics of office-based delivery sites that plan to hire more staff (787 respondents) Category Plan to Hire Additional Medical Staff Number of Column Row Sites Percent a Percent Do Not Plan to Hire Additional Medical Staff Number of Column Row Sites Percent a Percent Overall Parish Jefferson Orleans Plaquemines St Bernard Physician-based delivery site Yes No Service category Primary Care Medical Specialty Behavioral Health Other a May not total 100 percent due to rounding Summary of Table 8 The majority of delivery sites do not plan to hire additional medical staff within the next year (61.9%). Considerably less than half of the sites in Jefferson, Orleans, and Plaquemines parishes plan to hire additional medical staff, while six of seven (85.7%) responding sites in St Bernard Parish plan to hire additional medical staff. Considerably more physician-based sites are 12

16 planning to hire additional medical staff than non-physician-based sites (43.4% and 25.7%, respectively). By service category, other and primary care have the highest proportion of sites that are planning to hire additional medical staff (49.0% and 47.8%, respectively). Table 9: Characteristics of office-based delivery sites that use electronic medical records (779 respondents) Category Uses Electronic Medical Records Number of Sites Column Percent a Row Percent Does Not Use Electronic Medical Records Number of Column Row Sites Percent a Percent Overall Parish Jefferson Orleans Plaquemines St Bernard Physician-based delivery site Yes No Service category Primary Care Medical Specialty Behavioral Health Other a May not total 100 percent due to rounding Summary of Table 9 Survey respondents were asked to identify whether or not this practice currently uses electronic medical records (EMR). Because no definition of what constitutes EMR was provided, sites in both response categories might have electronic systems with varying capabilities. Overall, 14.2 percent of delivery sites in greater New Orleans currently use electronic medical records. The proportion of sites using EMR is roughly equal across parishes. A marginally higher proportion of physician-based sites use EMR as compared with non-physician-based sites (14.6% and 13.5%, respectively). By service category, the proportion of sites that currently use EMR is highest among other and primary care sites (18.2% and 15.2%, respectively), while the proportion is lowest among behavioral health sites (9.6%). 13

17 Table 10: Change in patient volume at office-based delivery site as compared with before Hurricane Katrina (769 respondents) Category Number of Sites Higher Patient Volume Lower Patient Volume No Change in Patient Volume Column Row Number of Column Row Number of Column Percent a Percent Sites Percent a Percent Sites Percent a Row Percent Overall Parish Jefferson Orleans Plaquemines St Bernard Physician-based delivery site Yes No Service category Primary Care Medical Specialty Behavioral Health Other a May not total 100 percent due to rounding Summary of Table 10 Because Table 10 compares patient volumes before and after Hurricane Katrina, the responses are limited to those sites that were operating before Katrina. With no information about the number and distribution of delivery sites that operated before Katrina, it is not possible to know if these responses are representative of all sites that operated before Katrina. 14

18 Overall, slightly fewer than half of the delivery sites (46.0%) that operated before Katrina reported a decrease in patient volume since the storm. The decrease in patient volume is most acute in St Bernard and Orleans parishes (85.7% and 59.5%, respectively). Meanwhile, in Jefferson Parish, only marginally more sites identified a decrease in patient volume rather than those that identified an increase (38.8 and 32.2, respectively). Among physician-based sites, the proportion that identified a decrease in patient volume was higher than the proportion among non-physician-based sites (48.5% and 40.4%, respectively). Between service categories, the highest proportion of sites that identified a decrease in patient volume occurred among other and primary care sites (50.6% and 50.3%, respectively) that had been operating before the storm and flood. Table 11: Annualized patient encounters at office-based delivery sites by parish and by service category (724 respondents) Category Patient Encounters Percent a Overall 4,220, Parish Jefferson 2,908, Orleans 1,251, Plaquemines 41, St Bernard 19, Physician-based delivery site Yes 3,536, No 684, Service category Primary Care 1,447, Medical Specialty 1,067, Behavioral Health 466, Other 1,238, a May not total 100 percent due to rounding Summary of Table 11 Survey respondents provided the number of patient encounters in the past month; these figures have been annualized in Table 11. These results have not been adjusted to account for nonresponse. Thus, the figures in Table 11 represent utilization at respondent delivery sites only. An estimated 4.2 million patient encounters took place at survey respondent delivery sites during a one year period. The large majority of these encounters (68.9%) occurred in Jefferson Parish 15

19 while Orleans Parish accounted for less than one-third (29.6%) and Plaquemines and St Bernard parishes together account for 1.5 percent of the region s patient encounters. Physician-based delivery sites provided more than five times the number of patient encounters as non-physicianbased sites (3.5 million vs. 0.6 million, respectively). Primary care delivery sites accounted for slightly more encounters than medical specialty and other delivery sites (34.3% vs. 25.3% and 29.4%, respectively) while behavioral health sites registered the fewest patient encounters (11.1%) at just less than half a million. 16

20 Table 12: Annualized patient encounters at office-based delivery sites by insurance type (699 respondents) Private Insurance Medicare Medicaid Uninsured Category Row Pct a Col Pct b Row Pct a Col Pct b Row Pct a Col Pct b Encounters Encounters Encounters Encounters Row Pct a Col Pct b Total Visits Overall 1,676, ,070, , , ,065,972 Parish Jefferson 1,221, , , , ,812,177 Orleans 432, , , , ,192,800 Plaquemines 14, , , , ,844 St Bernard 7, , , , ,152 Physician-based delivery site Yes 1,369, , , , ,423,084 No 306, , , , ,888 Service category Primary Care 526, , , , ,400,640 Medical Specialty 452, , , , ,006,105 Behavioral Health 146, , , , ,868 Other 550, , , , ,218,360 b Row Pct=Row Percent; may not total 100 percent due to rounding b Col Pct=Column Percent; may not total 100 percent due to rounding Summary of Table 12 Survey respondents provided the number of patient encounters in the past month; these figures have been annualized in Table 12. Encounter totals differ from those reported in Table 11 due to missing information on patients health insurance type for some delivery sites. 17

21 Overall, private insurance patients account for 41.2 percent of the encounters in the region, followed by 26.3 percent for Medicare patients, 15.1 percent for Medicaid patients, and 17.4 percent for uninsured patients. Jefferson Parish has the highest proportion of encounters by insured and Medicare patients (43.4% and 28.1%, respectively) while its burden of encounters by uninsured patients is the lowest of the four parishes (13.6%). The burden of encounters by uninsured patients is particularly high in St Bernard (37.2%) and Orleans (25.8%). Physicianbased delivery sites have a lower proportion of encounters from privately insured and Medicaid patients than non-physician-based sites. By service category, primary care and behavioral health sites have the lowest proportion of encounters by privately insured patients (37.6% and 33.1%, respectively) and the highest proportion of encounters by Medicaid (25.5% and 12.1%, respectively) and uninsured patients (18.0% and 44.3%, respectively). It is possible that some of the patients that were identified as uninsured by behavioral health sites have health insurance that does not cover mental healthrelated services. This might explain the relatively high proportion of encounters by uninsured patients at behavioral health sites (44.3%). STUDY LIMITATIONS The survey methods were carefully designed to minimize common sources of error and bias; however, the following limitations should be considered as they may affect the accuracy and interpretation of the reported results. Hospital emergency and outpatient departments were not surveyed. Therefore, the survey results are only representative of clinical or office-based delivery sites. As a result, to the extent that patients are receiving primary care in these hospital-based settings, and primary care providers are working in these settings, the survey results are likely to underestimate the total workforce and patient volume in ambulatory care in greater New Orleans. The survey questionnaire did not identify a category for resident physicians on staff. Respondents were asked not to include resident physician full time equivalents among any other physician category. Because resident physicians are licensed clinical providers and contribute substantially to workforce capacity, the survey results are likely to underestimate the total workforce in clinical or office-based delivery sites. In some cases, respondents recorded the number of providers in a given category rather than the number of full time equivalents. To reduce this source of error in reported full time equivalent providers, data collectors conducted follow-up telephone interviews with the majority of respondents during which they confirmed or corrected staffing information. In several tables, the estimated number of patient visits in the past month provided by survey respondents was annualized to produce a more useful indicator. Intensity of health care utilization may follow a season pattern due to the spread of common illnesses, as in the case of flu season, or due to insurance coverage, as in the case of annual deductibles and visits limits. Therefore, if the number of patient visits that occurred at the delivery site in the past month did 18

22 not approximate the monthly average for the year, then the annualized indicators might be overor underestimated. Because no single database could be identified to serve as a comprehensive source of the population for this study, it is possible that some eligible delivery sites in the region were not included in the survey population. Because of the use of several directories and extensive canvassing, the number of delivery sites that were missed by the study is likely to be very small. However, if sites were missed, then the study results are likely to underestimate the total workforce in clinical or office-based delivery sites. Moreover, if sites that were missed differ in important characteristics from the sites that were identified, then this may represent a limited but important source of bias. DISCUSSION This survey is the outcome of an attempt to assess the capacity of the ambulatory care workforce in the greater New Orleans area, post-katrina, to provide medical and behavioral health care to the resident population. The approach to identifying and surveying the existing service delivery sites was extensive and thorough, resulting in the most comprehensive and accurate identification of these sites to date. The survey response rate of over 70 percent is high by any standard. Therefore, the results of this survey are likely to provide an accurate picture of the health care workforce and service delivery sites in the region. This report only presents the data from the survey. No attempt has been made to draw conclusions regarding the adequacy of the available ambulatory care workforce in the region. There are two principal reasons for this approach. The first reason is that data on the size and characteristics of the current population of the region are uncertain. Population estimates for the summer of 2007 have not yet been released by the U.S. Census Bureau at the time of this report. Without reliable information about the region s population, it is problematic to determine if the current workforce is adequate to meet its health care needs. Two issues exacerbate this population problem: (1) there is limited current information about the health status and needs of the population and (2) there is little information about the geographic distribution of the population in relation to the location of the region s health care delivery sites. In regards to the first point, the 2006 Louisiana Health and Population Survey 10 and the Kaiser Family Foundation Post-Katrina Baseline Survey 4 provide a wealth of information about the health status of the region s population; however, both surveys were almost a year old at the time of this survey. In regards to the second point, impediments, such as distance to the delivery site, lack of public transportation, and travel time, have been identified as important potential barriers to health care access. Available methods for assessing the accessibility of health care rely on geographically detailed population data, for which post-katrina population estimates by the Census Bureau have thus far been insufficient. While the relative supply of physicians may have returned to pre-katrina levels, 9 many questions remain about the geographic, financial, and other potential barriers that might limit access to health care by the population at need. 19

23 The second reason that this report does not draw conclusions about the adequacy of the available ambulatory care workforce in the region concerns the lack of appropriate comparison data from national or regional workforce assessments. Most workforce assessments examine numbers of providers, rather than FTEs. There are good reasons for using FTEs, as this survey does most importantly that FTEs provide a more accurate assessment of the time providers spend on actual patient care. Nevertheless, no straightforward means exists for converting the number of FTEs to providers. Despite these issues, this survey represents the best estimate of the post-katrina ambulatory care workforce to date, based on responses from over 70 percent of the identified delivery sites, and it also provides some useful information regarding other aspects of health care delivery in the region. For example, there are twice as many physician-based delivery sites in Jefferson parish than in Orleans parish. In addition, the adoption of electronic medical records among all sites appears to be comparable to national averages, and patient volumes have shifted considerably from Orleans to Jefferson parish post-katrina. Many Orleans Parish delivery sites plan to hire medical staff in the next year. Orleans, Plaquemines, and St Bernard parishes face a considerably higher burden of encounters by patients without health insurance, for which these delivery sites are presumably un- or under-compensated. The strength of this survey is that it provides a snapshot of the characteristics of the workforce and delivery sites in the region during the time of the survey. The potential to periodically repeat this survey presents an important opportunity to identify trends in the recovery of the ambulatory health care system in greater New Orleans. AUTHORS AND ACKNOWLEDGEMENTS The Authors Gregory Stone, MS, conducts health systems planning at the Louisiana Public Health Institute. Mr. Stone designed the survey methodology, supervised implementation, and analyzed the data. Clayton Williams, MPH, is the Director of Health Systems Development at the Louisiana Public Health Institute. Mr. Williams defined the goals, conceptualized the methods, and supervised the project. Claudia Campbell, PhD, is a professor in the Department of Health Systems Management at Tulane University School of Public Health and Tropical Medicine. Dr. Campbell reviewed the data collection process and advised on the selection of indicators for reporting, methods of analysis, and interpretation of results. Mark Diana, MBA, MSIS, PhD, is an assistant professor in the Department of Health Systems Management at Tulane University School of Public Health and Tropical Medicine. Dr. Diana reviewed the data collection process and advised on the selection of indicators for reporting, methods of analysis, and interpretation of results. 20

24 Kara Wayman, MPH, is a Program Manager for Health Systems Development at the Louisiana Public Health Institute. Ms. Wayman managed the staffing and daily operations of the survey implementation. Acknowledgements Appreciation is expressed to the following institutions and individuals for their help in designing, implementing, and reporting this study: Louisiana Public Health Institute Susan Bergson, Lisanne Brown, Tung Ly, and Robert Habans Data collection team: Lani Clark, Yordanka Koleva, Dorothy Hayes, Steve Pinell, Will Lucas, Nickie Frederick, Betty Williams, and Charlene Grant Data processing team: Maura Reynolds, Heather Parker, and Meredith Dudley Louisiana Department of Health and Hospitals Kristy Nichols, Michael Dailey, and Anita Milling Tulane University Health Sciences Center Anthony Keck Appreciation is also expressed to the following institutions for their financial support of this project: Baptist Community Ministries HCA, Inc. The Fry Foundation ABOUT LOUISIANA PUBLIC HEALTH INSTITUTE The Louisiana Public Health Institute (LPHI) was established in 1997 and is one of 25 public health institutes across the country. LPHI is a private non-profit organization (501(c) 3) that works statewide in coordination with government, academia, and other health and human service agencies to advance the health and well being of the people of Louisiana through public-private partnering at the community, parish, and state levels. The initiatives of LPHI improve the health of Louisiana s people and communities, and provide information for decision making and policy development. Current activities employ a community-based systems approach to improving access to care, environmental tobacco policy change, and enhancing the capacity of communities to address critical health issues. Our focus is on promoting community-oriented solutions to public health issues and serving as the broker for traditional and non-traditional partnerships. 21

25 LPHI connects people, ideas, organizations, and resources to bring about positive change. We re about creating the space for innovative change to happen in Louisiana. For more information about LPHI s programs and partnerships, go to REFERENCES 1. Current Housing Unit Damage Estimates: Hurricanes Katrina, Rita, and Wilma. Washington, D.C.: Office of Policy Development and Research, U.S. Department of Housing and Urban Development; February 12, Williamson D. Study shows Hurricane Katrina affected 20,000 physicians, up to 6,000 may have been displaced. Chapel Hill: University of North Carolina, Chapel Hill; September 15, Madamala K, Campbell CR, Hsu EB, Hsieh Y-H, James J. Characteristics of Physician Relocation Following Hurricane Katrina. Disaster Medicine and Public Health Preparedness. 2007;1(1): Health Challenges for the People of New Orleans: The Kaiser Post-Katrina Baseline Survey. Menlo Park, CA: The Henry J. Kaiser Family Foundation; Anderson P. Hurricane Katrina survivors face increasing mental health problems. Medscape Medical News [ Accessed November 21, New Orleans' Parishes Top Nation in Population Growth Rate. Accessed March 20, Rudowitz R, Rowland D, Shartzer A. Health care in New Orleans before and after Hurricane Katrina. Health Aff (Millwood). Sep-Oct 2006;25(5):w Migration Patterns: Estimates of Parish Level Migrations Due to Hurricanes Katrina and Rita. Baton Rouge: Louisiana Recovery Authority; Griggs T. Report: N.O. patients struggle for care. The Advocate. October 1, 2007, 2007: 10A Louisiana Health and Population Survey. Accessed December 17,

26 APPENDIX A: SURVEY QUESTIONNAIRE 23

27 + + LOUISIANA DEPARTMENT OF HEALTH AND HOSPITALS GREATER NEW ORLEANS HEALTHCARE SERVICE CAPACITY ASSESSMENT A survey assessment of the healthcare service capacity and adoption of electronic medical records in the Greater New Orleans area INSTRUCTIONS The person completing this form should be someone with strong knowledge of the administration, services, and medical staffing at this practice. If have questions or need help, please call (504) PLEASE COMPLETE THIS FORM IN BLACK INK. PRINT NEATLY IN CAPITAL LETTERS AS SHOWN. L O U I S I A N A MARK THE APPROPRIATE BOX WITH AN X. IF YOU MAKE A MISTAKE, SHADE OUT THE BOX AND MARK THE CORRECT ONE. TODAY S DATE (MM/DD/YY) IMPORTANT: The questions in this survey refer ONLY to the address of the practice location that you indicate below. If this practice has multiple locations, a similar questionnaire will be sent to each practice location separately. Please contact us if you have questions. Please provide the following contact information for this practice location. NAME OF PRACTICE STREET NUMBER STREET NAME ADDRESS LINE 2 (for SUITE NUMBER or OFFICE NUMBER or OTHER) CITY ZIP CODE TELEPHONE FAX WEBSITE Which of the following best describes the person who is completing this form? (Mark all that apply) CEO/CFO Nurse Did this practice relocate due to hurricanes Katrina or Rita? Yes No Clinic administrator Office clerk Other, please specify: Physician Volunteer If YES, in what ZIP CODE was this practice previously located? FOR OFFICE USE ONLY

28 Which of the following best describes this practice? (Mark all that apply) Free standing, independent medical group Medical group component of integrated delivery system (IDS) Federally Qualified Health Center, Community Health Center or similar practice Medical school faculty practice plan Public Other non-profit organization (501c3) Other academic practice Other, please specify: 2. Which of the following best describes the majority owner of this practice? Government Hospital / integrated delivery system (IDS) Management Services Organization (MSO) or Physician Practice Management Company (PPMC) Physicians University or medical school Organizational component of an academic medical institution Non-profit organization (501c3) Other, please specify: 3. Does this practice offer bilingual services in Spanish? Yes No 4. Patient volume Estimate the number of active patients at this practice. (this is the number of individual patients seen at this practice in the past 18 months) Estimate the number of patient visits in the past month. 5. How many hours per day does this practice provide patient services? Monday Tuesday Wednesday Thursday Friday Saturday Sunday 6. When are the next three available appointments for a NEW patient? (Mark one for each appointment) 1 st appointment 1 week 2 weeks 3 weeks 4 weeks 5 weeks 6 weeks More than 6 weeks 2 nd appointment 3 rd appointment 7. Estimate the percent of this practice that is made up of patients in the following age groups: Age Group % 0 18 years years 8. Approximately what percent of the patients at this practice have the following types of insurance coverage? Health Insurance % Medicare Medicaid years Private Insurance 65 years and older No insurance FOR OFFICE USE ONLY

29 Was this practice affected in any of the following ways by Hurricane Katrina? (Mark all that apply) Received flooding Received damage from wind or rain Was closed for less than 6 months Was closed for 6 months or more Loss or damage to medical equipment Loss or damage to patient medical records Loss or damage to computer hardware Lost staff Other, please specify: 10. How is this practice operating now as compared with before Hurricane Katrina? More Less No change Number of days / hours open per week Services offered Patient volume Staff 11. Would this practice be interested in applying for a low interest loan to help restore or expand clinical services? Yes No 12. At what reimbursement level would this practice be agreeable to accepting and treating currently uninsured patients? 100% Medicare 90% Medicare Medicaid Don t know Not at any reimbursement level Other, please specify: 13. Within the next year, does this practice plan to (Mark all that apply) Hire additional medical staff Expand clinical services Open an additional facility Relocate current facility Change ownership Undergo major upgrades or renovation Other, please specify: USE OF ELECTRONIC MEDICAL RECORDS (EMR) AT THIS PRACTICE 14. Does this practice currently use electronic medical records (EMR)? Indicate the year this practice began using EMR Yes Indicate the manufacturer of the EMR system at this practice No Does this practice plan to adopt EMR? Yes, within 1 year Yes, but not within 1 year No, there are no plans to adopt EMR at this time 15. How many times has this practice considered or tried to adopt an EMR and abandoned it? FOR OFFICE USE ONLY

30 Which of the following EMR / practice management capabilities are currently in use or are being considered for use? (Mark all that apply) EMR / PRACTICE MANAGEMENT Scheduling (one site only) Scheduling (at other clinics within organization) Scheduling (at other sites) Registration (e.g. demographics, insurance) Billing (use of Superbill) Electronic claims submission Relay and track phone calls In Use Plan to Go-Live within 1 year 17. Which of the following EMR system capabilities are currently in use or are being considered for use? (Mark all that apply) Plan to EMR In Use Go-Live within 1 year Presenting complaint Problem list Past medical history Physical exam / review of systems Visit or encounter notes Evaluation and Management (E/M) coding Procedures E-prescribing Medication list Medications alerts for contraindications Print and track orders (labs, procedures, immunizations, referrals, consults) Lab results Radiology / imaging results Review reports / notes Messaging system Reminders for follow up procedures 18. Please indicate how each potential barrier affects this practice s decision to continue or expand the use of EMR. If this practice does not currently use EMR, please indicate how much each barrier contributes to why this practice does not use EMR. Productivity Lack of time to acquire, implement EMR Entering data into computer can be cumbersome No time to learn how to use EMR EMR is difficult to use Disrupts workflow and/or office s physical layout Temporary loss of productivity and/or revenue during implementation Time required to enter paper records into EMR Takes time away from providing patient care Financial Initial investment for hardware and software Maintenance costs Technical Products available do not meet practice s needs Staff lacks basic computer knowledge Temporary loss of access to patient records if system fails Patients Privacy/confidentiality concerns Patient resistance or do not want their physician to use EMR Major Barrier Minor Barrier Not a Barrier FOR OFFICE USE ONLY

31 Does this practice currently use a Registry or Disease Management system? Yes If YES, which of the following conditions are followed? (Mark all that apply) Report within Report to DISEASE MANAGEMENT No Practice State Diabetes Coronary Artery Disease Hypertension Heart Failure Depression Asthma Cancer Communicable Diseases Immunizations CLINICAL SERVICES OFFERED AT THIS PRACTICE 20. Which of the following services does this practice offer? Is there a discount for the uninsured? (Mark all that apply) PRIMARY MEDICAL CARE SERVICES On Site By Referral General Primary Medical Care (other than listed below) Diagnostic Laboratory (technical component) Diagnostic X-Ray Procedures (technical component) Diagnostic Tests/Screenings (professional component) Emergency medical services 24-hour coverage Family Planning Early Periodic Screening, Diagnosis and Treatment (EPSDT) HIV counseling and testing Testing for Blood Lead Levels Immunizations OBSTETRICAL AND GYNECOLOGICAL CARE Gynecological Care Prenatal care DENTAL CARE SERVICES Dental Care - Preventive MENTAL HEALTH/SUBSTANCE ABUSE SERVICES Mental Health Treatment/Counseling 24-hour Crisis Intervention/Counseling Substance Abuse Treatment/Counseling OTHER PROFESSIONAL SERVICES Pharmacy - Licensed Pharmacy staffed by Registered Pharmacist Pharmacy - Provider Dispensing Nursing home and assisted-living placement PREVENTIVE SERVICES Smoking cessation program Glycosylated hemoglobin measurement for patients with diabetes Blood pressure monitoring Discount for Uninsured FOR OFFICE USE ONLY

32 + + STAFFING AT THIS PRACTICE 21. Indicate the total number of Full Time Equivalents (FTEs) for each service category at this practice location. DO NOT include resident physicians. Example: How to calculate FTEs Family Practitioners Hrs/week FTE Physician Physician Physician Total What is a Full Time Equivalent (FTE)? FTE is a unit of measure which is equal to one full time (40 hours per week) position. So, a physician who works 10 hours per week at this practice is said to constitute 0.25 FTE ( = ). Personnel by Major Service Category Primary Care Current FTEs FTEs Needed Personnel by Major Service Category Other Medical and Surgical Specialties Current FTEs FTEs Needed Family Practitioners General Surgeons General Practitioners Orthopedic Surgeons Internists Cardiologists Obstetrician / Gynecologists Dermatologists Pediatricians Urologists Pediatric Sub- Specialist Neurologists Primary Care Specialists (including geriatrics, sports, preventative) Ophthalmologists Nurse Practitioners Otolaryngologists Physician Assistants Other Medical Specialists Certified Nurse Midwives Other Surgical Specialists Registered Nurses Behavioral Health Other Medical Personnel Laboratory Personnel and Technicians Dentists Psychiatrists Other Licensed Mental Health Providers Substance Abuse Services Other Licensed Dental Providers Pharmacists Administrative Staff Other Programs and Services, specify: FOR OFFICE USE ONLY

33 + + MAILING INSTRUCTIONS Before returning this form by mail please be sure that you have Provided the practice s complete address and contact information on the front page. Answered every question. Provided comments or questions in the space provided below. Then Put the questionnaire into the postage-paid return envelope. If the envelope has been misplaced, please mail the questionnaire to: GNO Healthcare Service Capacity Assessment c/o Louisiana Public Health Institute 1515 Poydras St, Suite 1200 New Orleans, LA Or Call us at (504) and we ll pick it up at your convenience! Thank you for participating! If you have comments please write them in the space provided below Or Send them in an to popest-la@lphi.org Comments FOR OFFICE USE ONLY

34 + + LOUISIANA DEPARTMENT OF HEALTH AND HOSPITALS GREATER NEW ORLEANS HEALTHCARE SERVICE CAPACITY ASSESSMENT A survey assessment of the healthcare service capacity and adoption of electronic medical records in the Greater New Orleans area Thank you for your participation! Information that you provide about this practice will be combined with information from other practices to assess the healthcare service capacity and electronic medical records (EMR) usage in your region. For more information about this survey, please contact Louisiana Public Health Institute 1515 Poydras St, Suite 1200 New Orleans, LA (504) This project is conducted by a collaboration of the following organizations: Health Sciences Center FOR OFFICE USE ONLY

35 APPENDIX B: MAPS OF SERVICE DELIVERY SITE LOCATIONS 32

36 33

37 34

38 35

39 Louisiana Public Health Institute is a private non-profit organization that works statewide in coordination with government, academia, and other health and human service agencies to advance the health and well being of the people of Louisiana through publicprivate partnering at the community, parish, and state levels. Louisiana Public Health Institute 1515 Poydras Street, Suite 1200 New Orleans, LA Phone: (504) Fax: (504)