UBC Faculty Research and Publications

Developing an information system to identify and describe physicians in clinical practice in British… Watson, Diane E.; Peterson, Sandra; Young, Ella; Bogdanovic, Bogdan Aug 31, 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-Watson_D_et_al_Info_system.pdf [ 661.21kB ]
Metadata
JSON: 52383-1.0048415.json
JSON-LD: 52383-1.0048415-ld.json
RDF/XML (Pretty): 52383-1.0048415-rdf.xml
RDF/JSON: 52383-1.0048415-rdf.json
Turtle: 52383-1.0048415-turtle.txt
N-Triples: 52383-1.0048415-rdf-ntriples.txt
Original Record: 52383-1.0048415-source.json
Full Text
52383-1.0048415-fulltext.txt
Citation
52383-1.0048415.ris

Full Text

Developing an  Information System  to Identify and  Describe Physicians  in Clinical Practice  in British Columbia 1996/97-2004/05August 2006Diane E Watson PhD MBASandra Peterson MScElla Young MHABogdan Bogdanovic BComm BAAN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 2Library and Archives Canada Cataloguing in PublicationDeveloping an information system to identify and describe physicians in clinical practice in British Columbia (1996/97-2004/05) [electronic resource] / Diane E Watson ... [et al.].ISBN 1-897085-05-2        1. Physicians (General practice)--British Columbia.  2. Group medicalpractice--British Columbia.  3. Primary health care--British Columbia. I. Watson, Diane II. University of British Columbia. Centrefor Health Services and Policy Research  III. Title.R729.5.G4M334 2006                    610                              C2006-901699-2  3AUGUST 2006Table of ContentsAbout CHSPR ....................................................................... iAcknowledgements ............................................................. iiExecutive Summary ............................................................. iiiIntroduction .......................................................................... 1Methods ................................................................................ 2  Findings ................................................................................ 5Discussion ............................................................................ 8Conclusion ............................................................................ 9References ............................................................................ 10Appendix ISource Data File Descriptions ............................................. 12Appendix IIPhysicians Who Did and Did Not Receive MedicalServices Plan Fee-For-Service Payments in 1996/97 and 2000/01 ............................................................. 14AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 4About CHSPRThe Centre for Health Services and Policy Research (CHSPR) is an independent research centre based at the University of British Columbia. CHSPR’s mission is to stimulate scientifi c enquiry into issues of health in population groups, and ways in which health services can best be organized, funded and delivered. Our researchers carry out a diverse program of applied health services and population health research under this agenda.CHSPR aims to contribute to the improvement of population health by ensuring our research is relevant to contemporary health policy concerns and by working closely with decision makers to actively translate research fi ndings into policy options. Our researchers are active participants in many policy-making forums and provide advice and assistance to both government and non-govern-ment organizations in British Columbia (BC), Canada and abroad. CHSPR receives core funding from the BC Ministry of Health Services to support research with a direct role in informing policy decision-making and evaluating health care reform, and to enable the ongoing development of the BC Linked Health Database. Our researchers are also funded by com-petitive external grants from provincial, national and international funding agencies. Much of CHSPR’s research is made possible through the BC Linked Health Database, a valuable resource of data relating to the encounters of BC residents with various health care and other sys-tems in the province. These data are used in an anonymized form for applied health services and population health research deemed to be in the public interest.CHSPR has developed strict policies and procedures to protect the confi dentiality and security of these data holdings and fully complies with all legislative acts governing the protection and use of sensitive information. CHSPR has over 30 years of experience in handling data from the BC Ministry of Health and other professional bodies, and acts as the access point for researchers wishing to use these data for research in the public interest.For more information about CHSPR, please visit www.chspr.ubc.ca.i 5AUGUST 2006AcknowledgementsThe conceptual and methodological advancements used to create the anonymized physician registry described in this report were guided by contributions from Diane Watson, Bogdan Bogdanovic and Nino Pagliccia. Sandra Peterson completed the programming required to establish the registry, and Sherin Rahim-Jamal helped coordinate activities in the early stages of the project. Ella Young assisted Diane Watson in writing the manuscript, and all authors contributed to, and approved, its content. We would also like to acknowledge colleagues who provided feedback on draft copies of this report.  The BC Ministry of Health provided funding, under the Primary Health Care Transition Fund, to support this project. The results and conclusions are those of the authors and no offi cial endorsement by the Ministry is intended or should be inferred. This project is part of a larger program of research conducted at the request of the Ministry.The BC Ministry of Health and the Behavioural Research Ethics Board of the University of British Columbia approved this project, and the College of Physicians and Surgeons of British Columbia approved the use of their data.iiAN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 6In 2001 and 2004, health human resources was identifi ed as a top priority by national consultations intended to illuminate areas where new research could inform health care policy and planning in Canada. Participants indi-cated that evidence relevant to workforce planning was increasingly needed to inform the actions and policies of governments, as well as educational, regulatory and pro-fessional bodies. A number of factors are driving the need for new evidence about the evolving nature and composi-tion of the health care workforce, including the practice patterns of physicians.This report describes methods established at the Cen-tre for Health Services and Policy Research (CHSPR) to develop and validate an anonymized registry of the physician workforce in clinical practice in Brit-ish Columbia. It outlines the protocols used to ensure the anonymity of providers, identifies key elements required in a physician registry in order to support population-based, provider-based and cohort analy-ses, compares physician supply estimates derived from single data sources to estimates drawn from multiple sources, and provides examples of how the registry can be used to support policy-relevant research.As the size of alternative payments for physician services grows, the usefulness of fee-for-service (FFS) billing data as a source of population-based information declines. How signifi cant a problem does this present in estimating physician supply? In 2004/05, 97 per cent of physicians in clinical practice could be identifi ed using FFS billing data. Of the remaining physicians (N=290) in clinical practice unaccounted for in FFS data, 69 per cent could be identi-fi ed using the Hospital Discharge Abstracts Database. Similar fi ndings were evident in 1996/97 and 2000/01. Th e combined use of FFS and hospital data now off ers the most effi  cient means of improving estimates of supply. Over time, underestimation in physician counts through methods that solely rely on FFS data will become ac-centuated and research analyses that rely on variables solely available in FFS data will become impracticable as: (a) the Ministry of Health increasingly uses alternative payments to fund service organizations and/or remuner-ate individual physicians; and (b) physicians increasingly seek remuneration solely from service organizations that receive alternative payments (i.e. block payments). Th ere is evidence that both trends are occurring, albeit gradually. Between 1996/97 and 2004/05, the percentage of clinical payments to physicians attributable to alternative funding of all types in BC increased from 6.8 per cent to 9.5 per cent. Th e percentage attributable to service organizations via block payments increased from 3.1 to six per cent of total expenditures on physician services.  Th e absence of encounter information about physicians and/or patients in alternative payment program (APP) databases introduces bias in supply-based analyses of FFS data, but this bias can be detected and measured through the use of the anonymized physician registry, which includes information about the degree to which each physician’s FFS data off ers a complete (or incomplete) picture of their clinical activities. Th e sole use of FFS billing data underestimates the work-force of physicians in clinical practice in British Columbia in 2004/05, although the problem is not yet substantial. But, temporal shift s in methods of payment with increases in the proportion of remuneration through APP suggest that the situation will get worse if current trends in fund-ing continue. Importantly, we do not assess the degree to which erosion of encounter data about physicians and their patients (previously available in FFS data but no longer available in the majority of new databases designed to track alternative payments) introduces measurement error into supply-based analyses regarding the practice patterns of physicians. Th is issue will be addressed in a forthcom-ing report. In essence, we can count most physicians, but we increasingly don’t know what services they provide and what services patients receive.Executive Summaryiii 1AUGUST 2006In 2001 and 2004, health human resources was identifi ed as a top priority in national consultations intended to illumi-nate areas where new research and knowledge translation activities could inform health care policy and planning in Canada. Participants indicated that evidence relevant to workforce planning was increasingly necessary to inform the actions and policies of governments, as well as educational, regulatory and professional bodies.1 A number of factors are driving this need for new evidence, including temporal shift s in workforce demographics and evolving practice patterns. Over the last decade, fee-for-service (FFS) billing data have increasingly been used to understand the supply and distribution of physicians, patterns of practice among these providers, and the use of physician services by Canadians. At the same time, there has been growth in alternative fund-ing of physicians, and in the number of databases that track those payments. Databases designed to track these payments vary within and between ministries of health—some include information about providers and patients, others do not. High quality evidence regarding the physician workforce and physician practice patterns requires reliable, complete and accurate information systems. Increasingly, our ability to de-scribe and analyze the physician workforce using traditional data sources such as College of Physicians and Surgeons of British Columbia (CPSBC) registration and FFS billing data could be eroding. For example, as physicians in specifi c spe-cialties increasingly expand or narrow the comprehensive-ness of their practice2, there is increasing concern regarding the validity of workforce analyses that rely solely on CPSBC registration data in describing these practices. Since pedia-tricians, geriatricians and other registered specialists might off er primary care, secondary care or a mix of both, informa-tion regarding registered specialty does not off er insights regarding the supply of primary care. Furthermore, as physi-cians increasingly receive remuneration through an array of funding arrangements now tracked in new databases, there is increasing concern regarding workforce analyses that rely solely on FFS billing data, since it may no longer represent Introductiona complete picture of the clinical activity undertaken by physicians. Over time, as College and FFS databases provide an incomplete picture of the nature and composition of the medical workforce, analyses generated from these sources may no longer be valid.   Th erefore, this report describes innovative methods used to: (a) develop and validate an anonymized registry of physicians deemed to be delivering clinical services; and (b) document the degree to which increases in the use of alternative fund-ing have altered our capacity to count the clinical physician workforce in British Columbia. It also describes protocols designed to ensure the anonymity of physicians, outlines key elements needed in a physician registry to support popula-tion-based, provider-based and cohort analyses, assesses variations in physician supply estimates when single versus multiple data sources are used, and off ers examples of how a physician registry can be used to support important, policy-relevant research. Th e results presented here rely on data from 1996/97, 2000/01 and 2004/05.      AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 2MethodsA two-stage process was used to develop an anonymized registry of the physician workforce in clinical practice in British Columbia. Th is project was approved by the BC Ministry of Health, the College of Physicians and Surgeons of British Columbia (CPSBC), and by an ethics committee at the University of British Columbia as part of a program of research related to primary health care. Stage One: Building the FoundationTh e foundation of the registry is a computer algorithm used to scan Medical Services Plan (MSP) practitioner fi les and CPSBC registration fi les to generate a parsimoni-ous set of non-identifying information regarding physi-cians. Physicians were selected at this stage if they were:(a) eligible to bill MSP and;(b)  registered with CPSBC and; (c)   residents of the province during the fi scal    period.* Th e MSP practitioner fi le is a historic database that records all practitioners ever eligible to submit claims for services. Th ese fi les are created by the Ministry of Health and in-clude information about physicians and other allied health professionals. Likewise, the CPSBC registration fi le is a historic database that records all physicians ever licensed to practice in British Columbia. Th e registration fi les are updated on an annual basis through self-report question-naires. All physicians are assigned a registration status such as full register, retired or expired. Historically, physician supply estimates have relied on the full register status code. To our knowledge, no one has cross-validated this fi eld with administrative billing data regarding service delivery in order to confi rm a physician is in clinical practice.Th e MSP practitioner fi le is created for fi scal periods and includes physicians eligible to bill until March 31 of a given year. By comparison, the CPSBC fi le is created by calen-dar year. Th is necessitates the use of two calendar years of CPSBC data to identify physicians recorded in a single fi s-cal year of MSP practitioner data. Th e fi rst calendar year of CPSBC data are required to identify physicians in the MSP practitioner fi le until December 31 in one calendar year, and the second calendar year of CPSBC data are required to identify physicians who emerge in the MSP practitioner fi le between December 31 and March 31. Stage Two: Determining Clinical PracticeTo identify physicians actively engaged in clinical practice, we created a computer algorithm that scanned an array of service and payment fi les to detect any records regarding clinical activity for the population of physicians identifi ed in stage one. Th ese service and payment fi les included: MSP fee-for-service (FFS) billings for provincial and out-of-province residents, the Hospital Discharge Abstracts Database, Primary Health Care Organizations data, and APP data related to individual physicians (e.g. salary and sessional payments). Th e BC Ministry of Health ensures that each physician has a unique identifi er common across all these data sources, and this identifi er enables computer algorithms to be used on anonymized fi les. Source fi les are described in more detail in Appendix I. *  In order to exclude those who meet these criteria but resided out-of-province, we limited the registry to physi-cians that have a postal code indicating residence in Brit-ish Columbia (starting with V) in both data sources. 3AUGUST 2006Figure 1: Data Sources Used to Create an Anonymized Physician RegistryPhysicians were included at this stage if they had billed for services in at least one of the following ways:(a)  have at least one billing record in one of the administrative fi les (MSP FFS for provincial and out-of-province residents or APP billing data)‡;  (b)  have at least one record in the Primary Health Care Organization data fi le;(c)  have at least one service record in the Hospital Discharge Abstract Database.Th e resulting anonymized physician registry contained the following variables: a unique identifi er, basic de-mographic information, registration (licensure) start and stop dates, geographic region of practice, physician specialty, type of practice§ and codes to identify in which data source the physician’s data appeared.Importantly, APP data in British Columbia includes records of payments made to individual physicians and to service organizations. Neither of these payment types specifi es the patients to who services were delivered. APP records relating to individual physicians can be, and were, assigned to them, but APP records relating to service agreements with organizations can only be attributed to the organization or service program. Th us, a physician would not be included in the registry as being engaged in clinical practice if they solely delivered clinical services through one of these APP funded organizations or service programs and never delivered clinical care that would gen-erate a record in any other database identifi ed in Figure 1. At the time of writing, researchers at CHSPR did not have data on alternative payments made through the Medi-cal On-Call Availability Program or the Rural Practice Programs or their predecessors. Th e Canadian Institute for Health Information (CIHI) reports that these programs accounted for $40 million and $163 million in expendi-tures in 2000/01 and 2002/03 respectively.‡ Th ese fi les include services funded by the Insurance Corporation of British Columbia (ICBC) and WorkSafeBC (Workers’ Compensation Board of BC). § Th e type of practice designation represents a physician’s functional specialty. It is a variable developed and com-puted by British Columbia’s Ministry of Health. One of the unique features of the FFS schedule in British Columbia is that billable fee items, with a small number of excep-tions, are assigned to a specialty group. A physician’s type of practice can be determined by identifying the specialty group under which most of their fee items are billed.  AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 4Comparing Estimates Using any single fi le or subset of fi les listed in Figure 1 in order to count the number of physicians in British Columbia or estimate the number of physicians engaged in clinical practice, could result in incomplete estimates. Insofar as those responsible for estimating workforce sup-ply use an array of methods and data sources, estimates will obviously vary. Th erefore, we assessed the diff erences in estimates of physician supply using diff erent subsets of data sources, or diff erent methodologies.  We compared workforce counts estimated with the newly created anonymized physician registry with those generated through MSP FFS fi les. To do this, we counted the number of physicians in the anonymized registry that only appear in the FFS data fi les and calculated the percentage these physicians represent of those in clini-cal practice (i.e. included in the anonymized physician registry). To identify the primary data source from which non-FFS physicians could otherwise be identifi ed, we counted the number of physicians not included in the MSP FFS payment fi le, and calculated the percentage of these physicians identifi ed through the use of additional service and payment fi les such as the Hospital Discharge Abstracts Database and APP fi les.Next, we compared estimates of physician workforce based on: (a) the anonymized physician registry; (b) MSP FFS fi les; (c) a report published by CIHI; and (d) a report published by the CPSBC. We evaluate physician counts, variation in estimates, and measures of change in esti-mates for 1996/97, 2000/01 and 2004/05. 5AUGUST 2006FindingsBetween 1996/97 and 2004/05, the percentage of clinical payments to physicians attributable to alternative funding of all types in British Columbia increased from 6.8 per cent to 9.5 per cent. Th e percentage attributable to service organizations via block payments increased from 3.1 to six per cent of total expenditures on physician services.  We identifi ed 8,558 registered physicians deemed to be in active clinical practice in British Columbia in 2004/05 (Figure 2). Th ese physicians accounted for 99 per cent of all medical expenditures in 2004/05 MSP FFS data. Using a similar algorithm, we identifi ed 7,534 and 7,822 physi-cians in clinical practice in 1996/97 and 2000/01 respec-tively.Figure 2: Two-Stage, Multi-Source Method to Count Physicians in Clinical Practice in BC in 2004/05Among the workforce of 8,558 physicians actively engaged in clinical practice in 2004/05, 8,268 (97%) could be identi-fi ed using MSP FFS fi les (Table 1). In the same year, an ad-ditional 290 physicians (3%) were identifi ed in other service fi les (Tables 1 and 2). Of these 290 physicians: ·  199 (69%) could be identifi ed as engaging in clini-cal practice in hospitals. Only 25 of these hospital-based physicians received remuneration through APP disbursements to individual physicians (e.g. salary, sessional). Th e remainder might be funded through, for example, APP payments to service organizations, or regional health authorities; and·  89 (31%) received APP funding directly, rather   than through a service organization.AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 6Table 1: Physicians Who Received MSP FFS Payment in 2004/05PTO Total 439 239 131 - 16 864- 58 3,908 159 1 5 265 2 10 2,171100%  8,268290 100%5%3%2%-0%10%-1%47%2%0%0%3%0%0%26%%PHCOHospMSP APPPTO Total 174 25----- 89---- 260%9%-----31%----1%%PHCOHospMSP APP0405 PTO Total 400 205 115 - 28 920 1 21 4,091 134 2 3 178- 1 1,397100%  7,496326 100%5%3%2%-0%12%0%0%55%2%0%0%2%-0%19%%PHCOHospMSP APPPTO Total 192 27----- 107-----59%8%-----33%-----%PHCOHospMSP APP 0001Total 403 153 122 1,216 4,126 94 892 188100%  7,194340 100%6%2%2%17%57%1%12%3%%PTOHospMSP APPTotal 184 118 1 30 7 -54%35%0%9%2%-%PTOHospMSP APP9697Table 2: Physicians Who Did Not Receive MSP FFS Payments in 2004/05PTO Total 439 239 131 - 16 864- 58 3,908 159 1 5 265 2 10 2,171100%  8,268290 100%5%3%2%-0%10%-1%47%2%0%0%3%0%0%26%%PHCOHospMSP APPPTO Total 1745----- 89---- 260%9%-----31%----1%%PHCOHospMSP AP0405 PTO Total 400 205 115 - 28 920 1 21 4,091 134 2 3 178- 1 1,397100%  7,496326 100%5%3%2%-0%12%0%0%55%2%0%0%2%-0%19%%PHCOHospMSP APPl192 27-----07-----598%-----33-----PHCOHospMSP APP0001Total 403 153 122 1,216 4,126 94 892 188100%  7,194340 100%6%2%2%17%57%1%12%3%%PTOHospMSP APPTotal 184 118 1 30 7 -54%35%0%9%2%-%PTOHospMSP APP9697Using only those physicians in the anonymized registry with MSP FFS payment data for supply estimates (Table 1) fails to account for several hundred practitioners who receive funding through other sources (Table 2). Be-tween 1996/97, 2000/01 and 2004/05, a similar number of physicians (N=340, 326, 290) in clinical practice were unaccounted for using FFS billing data, and a similar proportion of these ‘lost physicians’ (65%, 67%, 69%) could be identifi ed using the Hospital Discharge Abstracts Database (see Appendix II for tables relating to 1996/97 and 2000/01).Our results do indicate some variation in estimates of physician workforce counts attributable to the source and methods of estimation (Figure 3). If the anonymized physician registry is considered a reference standard, the exclusive use of the Medical Service Plan practitioner fi le and FFS payment data to identify physicians in clini-cal practice resulted in low estimates. Importantly, the degree of underestimation is consistent over a period of time when alternative funding increased. Th erefore, the proportion of physicians in clinical practice who did not receive FFS payments was stable despite increases in use of alternative funding options. Th ere are slight diff erences Note: MSP = Medical Services Plan fee-for-service pay-ment database for provincial residents, Hosp = Hospital Discharge Abstracts Database, APP = Alternative payment program data for physicians, PHCO = Primary Health Care Organizations, TO = Medical Services Pl n fee-for-service payment database for out-of-province residents. 7AUGUST 2006Table 2: Physicians Who Did Not Receive MSP FFS Payments in 2004/05in our estimates of supply based on exclusive use of the MSP practitioner fi le and FFS payment data and estimates created by the BC Ministry of Health using FFS payment data which are attributable to diff erences in methodology when using these data.If the anonymized physician registry is considered a refer-ence standard, CIHI estimates were similar in 1996/97, higher in 2000/01 and lower in 2004/05. Diff erences in estimates are due to data sources and methodology. CPSBC reports the highest estimates in all three fi scal periods, perhaps because it includes both physicians en-gaged in clinical practice and those who work exclusively 1 Numbers derived from BC Linked Health Database, BC Ministry of Health and CPSBC. Physicians who billed in   any billing file were included.‘96 ‘00 ‘04 ‘96 ‘00 ‘04AnonymizedPhysicianRegistryCPSBCRegistrationFileAPP data forphysiciansPrimaryHealth CareOrganizationsDatabaseHospitalDischargeAbstractsDatabaseMSPPractitionerFileMSP fee-for-service files for provincial and out-of-provinceresidentsBC MDs present in both CPS and MSP with billings1 in 2004/05N=8,558MDs present in both CPS and MSPn=14,772CPS only, not in MSPn=10,408MSP only, not in CPSn=10,669MDs with BC postal codes, both sourcesn=12,583No billings in 2004/05n=4,025(465 were full register)2000 CPSRegistrationFilen=21,3542000/01 MSPPractitionerFilen=22,0362005 CPSRegistrationFilen=25,1802004/05 MSPPractitionerFilen=25,441Stage IStage  II1At least one claim in MSP PIM, MSP PIM subset for out-of-province residents, PHCO, or APP physician-level data,  or at least one service record in the HSP-DAD2 Numbers derived from BC Linked Health Database, BC Ministry of Health. Only physicians who had billing records in the   MSP Payment Information Masterfile and who had BC postal codes in the MSP practitioner file were included.3 BC Ministry of Health, ‘MSP Information Resource Manual FFS Payment Statistics’ 1999-2000,  2001-2002,   and 2004-2005, p. 58, 16, and 17 respectively. 4 CIHI, 'Supply, Distribution and Migration of Canadian Physicians 2005' with a caveat noted for 2004: 'Data for the year   2004 do not reflect annual physician information provided by the College of Physicians and Surgeons of British Columbia.'5 CPSBC 'Annual Report' for 1997 and 2001, p. 28 and 35 respectively, and 2004 figure from personal communication   with Jian Liu of the College of Physicians and Surgeons of BC April, 2006. ‘96 ‘00 ‘04 ‘96 ‘00 ‘04CIHI4 CPSBC5Ministry of Healthfee-for-servicephysicians3CHSPR anonymizedphysician registry1CHSPR fee-for-servicephysicians2‘96 ‘00 ‘04Number of physicians7,0007,5008,0008,5009,000Figure 3: Estimates of Physician Workforce Counts (1996, 2000 and 2004)in management and administration, education, training and/or research. Th ere are substantive diff erences in physi-cian counts between CPSBC estimates in 2004 and those in the anonymized registry or reported by others. Despite diff erences in absolute counts of physicians in each time period, estimates of change in supply over the eight-year period were similar and ranged from ten per cent (Cana-dian Institute for Health Information) to 15 per cent (FFS payment fi les) (Figure 3).AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 8DiscussionOver the last number of years, FFS billing data have increasingly been used by the health care policy, plan-ning and practice communities to understand the supply and distribution of practitioners, the practice patterns of providers, and the use of services by Canadians. At the same time, there has been an increase in new funding and remuneration arrangements for physicians, and in the number and structure of databases that track these pay-ments. Th is report sheds light on the implication of these changes—regarding both the validity of supply estimates, and the validity of analyses intended to describe patterns of practice and use of services.   Th e implementation of new databases to track alterna-tive payments to service organizations has resulted in the loss of physician- and service-specifi c identifi ers. As a consequence, traditional FFS billing data incompletely estimates the supply of physicians, particularly those who solely receive remuneration from alternative payments. How signifi cant is the problem? In 2004/05, 97 per cent of the physicians in clinical practice in British Columbia could be identifi ed using FFS billing data. Of the remain-ing physicians (N=290) in clinical practice unaccounted for using fee-for-service billing data, 199 (69%) could be identifi ed through the use of the Hospital Discharge Abstracts Database. Th e remaining 31 per cent could be identifi ed through the use of APP data covering disburse-ments to individual physicians. In 1996/97, 2000/01 and 2004/05, a similar number of physicians (N=340, 326, 290) in clinical practice were unaccounted for using FFS billing data and a similar proportion of them (65%, 67%, 69%) could be identifi ed using the Hospital Discharge Abstracts Database. Th us, the combined use of FFS and hospital data off ers the most effi  cient means of improving estimates, given the structure and limitations of current information systems. Hospital data off er the opportunity to identify acute care physicians remunerated solely through APP service agreements or other sources. Insofar as community-based physicians are remunerated through APP disbursements to service organizations and receive no FFS income, the only method to identify them as being in clinical practice is through Primary Health Care Organization data. Physi-cians who work in these organizations receive blended funding of age-sex-morbidity adjusted capitation and FFS. Th ey submit not only FFS claims, but also patient-specifi c encounter records for services covered under capitation. Reporting is done using the claims system (Teleplan). Th e benefi ts of this arrangement include correctly identifying the active physician, as well as capturing diagnostic and service information about the population that they serve. It also demonstrates the feasibility of collecting encounter records through the FFS infrastructure. Underestimation in physician supply counts that rely solely on traditional FFS billing data will become accentuated over time as the Ministry of Health increasingly uses the APP program to fund service organizations and remuner-ate individual physicians, and as physicians increasingly seek remuneration solely from APP-funded organizations. Th ere is evidence to suggest that both are occurring. CIHI reports3 that clinical expenditures on physician services in British Columbia totaled $1.5 billion in 2000/01, of which 87.5 per cent was spent through MSP and 12.5 per cent through APP. Th e same report shows that expenditures rose to about $1.8 billion in 2003/04, of which 80.3 per cent was spent through MSP and 19.7 per cent through APP. Th is report has not been updated since that time.Failing to capture patient-specifi c encounter records (including diagnostic conditions responsible for a visit) in new databases designed to track alternative payments increases the risk of missing important temporal shift s in the practice patterns of physicians, in service use, and in treatment incidence-prevalence estimates for specifi c diseases. Importantly, APP data for individual physicians (e.g. salary and sessional) in British Columbia do not include information on type or date of service, diagnostic condition responsible for the visit, nor a patient identifi er. So while APP data could be used to count most but not all physicians and, potentially, to estimate their practice  9AUGUST 2006incomes, APP data are not useful for provider-based analyses regarding practice patterns or population-based analyses regarding patterns of service use. Th e absence of patient encounter information in APP data introduces bias in population-based service use analyses derived solely from FFS data. Th e magnitude of error among a geographic population or cohort will be directly related to the likelihood that they received ser-vices from specifi c APP-funded physicians and/or service organizations. For example, visit rates to pediatricians among a geographic population may be underestimated if pediatric service organizations in the community receive APP funding or if a high proportion of pediatricians are funded though alternative payments. CHSPR’s anony-mized physician registry can help estimate bias due to the former and the latter. Th e absence of physician-specifi c information regard-ing encounters funded through alternative payments increases the risk of missing important temporal shift s in the practice patterns of physicians and in service delivery patterns, and introduces bias in provider-based analyses intended to understand patterns of practice using FFS data. Th e anonymized physician registry can be used to estimate bias in analyses of the whole workforce or in cohorts of physicians by identifying the degree to which each physician’s FFS data off ers a complete picture of their clinical activities.ConclusionTh is report describes new methods established at the Centre for Health Services and Policy Research (CHSPR) to develop and validate an anonymized registry of the physician workforce in clinical practice in British Colum-bia. With the appropriate approvals, researchers could combine data elements from the registry with payment or service data to support more accurate analyses intended to describe physicians’ practice patterns (including volume, service type, group practice characteristics, continuity and specialty referrals) or a population’s use of physician services.Th e anonymized physician registry is one methodologi-cal solution to growing concerns regarding the validity of relying solely on traditional administrative data fi les to count and describe the physician workforce. It should also enable researchers to estimate bias in provider-based analyses regarding patterns of practice, and in population-based analyses regarding use of physician services that rely on encounter level information from FFS billing data. Researchers at CHSPR continue to develop and share methodologies intended to best combine FFS and alterna-tive payment data to estimate patterns of supply and use of physician services. Results from this work will continue to emerge in the coming months.   AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 101 Dault M, Lomas J, Barer M. on behalf of the Listening for Direction II partners. Listening for Direction II: National Consultation on Health Services and Policy Issues for 2004-2007. Final Report. Ottawa: Canadian Health Services Research Foundation; 2004. Available at: www.chsrf.ca/oth-er_documents/listening/pdf/LfD_II_Final_Report_e.pdf. Retrieved on November 29, 2005.2 Canadian Institute for Health Information. Evolving Role of Canada’s Family Physicians, 1992-2001. Ottawa (ON): Canadian Institute for Health Information; 2004.  3 Canadian Institute for Health Information. Th e Status of Alternative Payment Programs for Physicians in Canada, 2003-2004 and Preliminary Information for 2004-2005. Ottawa (ON): Canadian Institute for Health Information; 2006, p.5.4 McKendry R. Watson DE, Goertzen D, Reid RJ, Mooney D, Peterson S. Single and Group Practices Among Prima-ry Healthcare Physicians in British Columbia. Vancouver (BC): Centre for Health Services and Policy Research; 2006.5 Centre for Health Services and Policy Research. BC Linked Health Database. Available at: www.chspr.ubc.ca/data. Ac-cessed on July 25, 2005.6 Ibid, http://www.chspr.ubc.ca/data.7 BC Ministry of Health. Alternative Payments Program. Available at: www.healthservices.gov.bc.ca/pcb/app.html. Accessed September 22, 2005.8 Canadian Institute for Health Information. Th e Status of Alternative Payment Programs for Physicians in Canada, 2003-2004 and Preliminary Information for 2004-2005. Ottawa (ON): Canadian Institute for Health Information9 British Columbia Medical Association. Negotiations Information. How Physicians are Paid in BC – Fact Sheet. 2003.Available at: http://www.bcma.org/public/Negotia-tions_Information/HowPhysiciansGetPaidFactSheet.htm. Accessed on July 17, 2006.  10 British Columbia. Offi  ce of the Auditor General. Alter-native Payments to Physicians: A Program in Need of Change. Victoria (BC): Queen’s Printer for British Colum-bia; 2003.References 11AUGUST 2006AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 12Appendix I: Source Data File DescriptionsCollege of Physicians and Surgeons of British Columbia (CPSBC) FileA registration database created and maintained by the CPSBC, an organization responsible for licensing and regulating physicians in British Columbia. All practicing physicians must be registered, and complete a question-naire administered by the College each year. Th is fi le is used by CIHI and the CPSBC to generate physician counts for the province. For these analyses, we retained physicians with a valid British Columbia postal code.4  Medical Service Plan (MSP) Practitioner File An electronic data fi le created and maintained by the BC Ministry of Health. Contains information about medical practitioners that are registered with the College and who were eligible to bill MSP in 2000/01.  MSP Payment Information Master (PIM) FileAnnual, fi scal-year services provided to MSP-covered individuals by practitioners, billed to MSP and paid by MSP, WorkSafeBC, ICBC, the Midwife plan, or the reciprocal agreement. Th is fi le is used primarily by the BC Medical Association and the Ministry of Health to generate information about expenditures and counts of providers and consumers. Providers billing MSP are separated into three groups: physicians, supplementary benefi t practitioners, and out-of-province practitioners.5  Only the subset of services provided in BC is used for these analyses. Our analysis took into account retroac-tive adjustments or corrections by combining payment information regarding one visit. Th e results include a net-ted number of visits or services paid and a netted amount paid. Physicians that billed MSP for at least one netted service were selected for these analyses.MSP PIM Subset for Out-of-Province Residents (MSP-PTOOP)Records of residents of other provinces (except Quebec) who receive services in British Columbia.Hospital Discharge Abstracts DatabaseDischarges, transfers and deaths of in-patients and day surgery patients from acute care hospitals in BC.6 In-cludes diagnostic codes, hospital transfer and admittance codes, treatment regimens, and most responsible physi-cian numbers. Physicians who provided hospital services were identifi ed using the physician most responsible fi eld in each patient record.  Alternative Payment Plan (APP) FilesAt the time of writing, researchers at CHSPR had no data on alternative payments made through the Medical On-Call Availability Program or the Rural Practice Programs or their predecessors. CIHI reports that these programs accounted for $40 million and $163 million in expendi-tures in 2000/01 and 2002/03 respectively.APP Payments to Individual Physicians Th ese payments include salary, sessional and other individual payments to physicians such as callback, on call or northern allowance. Approximately 17 per cent of the province’s general practitioners were paid from this fund in 2000/01. However, no details have been provided about the services provided through this fund, such as who received the services or the types of services deliv-ered. Due to this lack of detail, only the dollar amounts and estimates regarding physicians involved were used in our analysis. 13AUGUST 2006APP Payments to Service Organizations (Service Agreement Data)An additional source of information on payments to physicians are APP program-level data. Th e Ministry of Health allocates most of these funds to health authorities, and some to other provincial health organizations. Th ese organizations then compensate physicians ‘in situations where FFS arrangements may not guarantee physicians the fi nancial support or stability to be able to provide needed care. Examples include teaching hospitals, community and hospital-based psychiatric services, and physician services in rural areas’.7  Th e site also states that the budget for alternative funding has increased steadily over the last decade and now represents about ten percent of the overall available amount for physician services. According to a recent CIHI report for fi scal 2003, alternative payments made up about 20 per cent of total physician compensa-tion in British Columbia.8 It is vital to consider this large and ever-growing fund in order to ensure that physicians, the services they provide, and the patients who receive those services, are properly accounted for.Service agreement payments accounted for about 58 per cent of the total APP budget in BC in 2000/01. Service organizations that receive allocations include hospitals, the BC Cancer Agency, the BC Centre for Disease Control, Riverview Hospital, Crown corporations, universities and private corporations. Physicians who receive these funds include emergency room physicians and pathologists.9 No detail has been provided at a system level about which particular physicians receive payments from this fund, or about who receives services paid for from this fund. As stated by the Offi  ce of the Auditor General of BC in 2005, ‘APP is undermined by weak or inadequate management systems’, and ‘(APP needs to) establish formal policies and procedures to ensure services are rendered in accordance with the agreements and all payments have proper ap-proval and are only made for services received.’10Primary Health Care Organizations (PHCO) FilePHCOs deliver patient care under a blended funding model. Unlike the fee-for-service only model, blended funding provides a per person-per day component that supports the delivery of common (‘core’) services to those patients who use the PHCO for the majority of their PHC services. Th ese payments appear as $0 amounts in MSP PIM FFS data since they are paid out of capitated funding. For patients who do not use the PHCO for the major-ity of their PHC services, the PHCO continues to submit fee-for-service (FFS) claims to MSP. Records in the PHCO fi le show up as shadow billings in the MSP PIM FFS fi le. Th us, PHCOs receive funds from the Ministry of Health to pay physicians directly and billings by PHCO physicians appear as a subset of the MSP PIM data. AN  I N FORMAT I ON  S YS T EM  TO  I D EN T I F Y  A ND  D E SCR I B E  P H YS I C I ANS  I N  C L I N I C A L  P RACT I C E  I N  B CCENTRE FOR HEALTH SERVICES AND POLICY RESEARCH 14Appendix II: Physicians Who Did and Did Not Receive Medical Services Plan Fee-For-Service Payments in 1996/97 and 2000/01Table 1: Physicians Who Received MSP FFS Payments in 1996/97Table 2: Physicians Who Did Not Receive MSP FFS Payments in 1996/97Table 3: Physicians Who Received MSP FFS Payments in 2000/01Table 4: Physicians Who Did Not Receive MSP FFS Payments in 2000/01PTO Total 439 239 131 - 16 864- 58 3,908 159 1 5 265 2 10 2,171100%  8,268290 100%5%3%2%-0%10%-1%47%2%0%0%3%0%0%26%%PHCOHospMSP APPPTO Total 174 25----- 89---- 260%9%-----31%----1%%PHCOHospMSP APP0405 PTO Total 400 205 115 - 28 920 1 21 4,091 134 2 3 178- 1 1,397100%  7,496326 100%5%3%2%-0%12%0%0%55%2%0%0%2%-0%19%%PHCOHospMSP APPPTO Total 192 27----- 107-----59%8%-----33%-----%PHCOHospMSP APP 0001Total 403 153 122 1,216 4,126 94 892 188100%  7,194340 100%6%2%2%17%57%1%12%3%%PTOHospMSP APPTotal 184 118 1 30 7 -54%35%0%9%2%-%PTOHospMSP APP9697PTO Total 439 239 131 - 16 864- 58 3,908 159 1 5 265 2 10 2,171100%  8,268290 100%5%3%2%-0%10%-1%47%2%0%0%3%0%0%26%%PHCOHospMSP APPPTO Total 1745----- 89---- 260%9%-----31%----1%%PHCOHospMSP AP0405PTO Total 400 205 115 - 28 920 1 21 4,091 134 2 3 178- 1 1,397100%  7,496326 100%5%3%2%-0%12%0%0%55%2%0%0%2%-0%19%%PHCOHospMSP APPt l 192 27----- 07-----59%8%-----33%-----PHCOHospMSP APP0001Total 403 153 122 1,216 4,126 94 892 188100%  7,194340 100%6%2%2%17%57%1%12%3%%PTOHospMSP APPTotal 184 118 1 30 7 -54%35%0%9%2%-%PTOHospMSP APP9697PTO Total 439 239 31 - 16 864- 58 3,908 159 1 5 265 2 10 2,171100%  8,268290 100%5%3%2%-0%10%-1%47%2%0%0%3%0%0%26%%PHCOHospMSP APPPTO Total 174 25----- 89---- 260%9%-----31%----1%%PHCOHospMSP APP0405 PTO Total 400 205 115 - 28 920 1 21 4,091 134 2 3 178- 1 1,397100%  7,496326 100%5%3%2%-0%12%0%0%55%2%0%0%2%-0%19%%PHCOHospMSP APPPTO Total 192 27----- 107-----59%8%-----33%-----%PHCOHospMSP APP 0001Total 403 153 122 1,216 4,126 94 892 188100%  7,194340 100%6%2%2%17%57%1%12%3%%PTOHospMSP APPTotal 184 118 1 30 7 -54%35%0%9%2%-%PTOHospMSP APP9697PTO Total 439 239 31 - 16 864- 58 3,908 159 1 5 265 2 10 2,171100%  8,268290 100%5%3%2%-0%10%-1%47%2%0%0%3%0%0%26%%PHCOHospMSP APPPTO Total 1745----- 89---- 260%9%-----31%----1%%PHCOHospMSP AP0405PTO Total 400 205 115 - 28 920 1 21 4,091 134 2 3 178- 1 1,397100%  7,496326 100%5%3%2%-0%12%0%0%55%2%0%0%2%-0%19%%PHCOHospMSP APPt l 192 27----- 07-----59%8%-----33%-----PHCOHospMSP APP0001Total 403 153 122 1,216 4,126 94 892 188100%  7,194340 100%6%2%2%17%57%1%12%3%%PTOHospMSP APPTotal 184 118 1 30 7 -54%35%0%9%2%-%PTOHospMSP APP9697Note: MSP = Medical Services Plan fee-for-service pay-ment database for provincial residents, Hosp = Hospital Discharge Abstracts Database, APP = Alternative payment program data for physicians, PHCO = Primary Health Care Organizations, PTO = Medical Services Plan fee-for-service payment database for out-of-province residents.Centre for Health Services and Policy ResearchThe University of British Columbia429 – 2194 Health Sciences MallVancouver, B.C. Canada  V6T 1Z3Tel:  604.822.4969Fax:  604.822.5690Email: enquire@chspr.ubc.cawww.chspr.ubc.caAdvancing world-class health services and policy research, training and data resources on issues that matter to CanadiansAUGUST 2006

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0048415/manifest

Comment

Related Items