Electronic Health/Medical – HCI 112

sing a data entry clerk to improve data quality in primary care electronic medical records: A pilot study-(Further reading- Optional)

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/230564870
Using a data entry clerk to improve data quality in primary care
electronic medical records: A pilot study
Article in The Journal of Innovation in Health Informatics · July 2011
DOI: 10.14236/jhi.v19i4.819 · Source: PubMed
CITATIONS
READS
17
721
11 authors, including:
Jan Barnsley
Babak Aliarzadeh
University of Toronto
University of Toronto
149 PUBLICATIONS 2,901 CITATIONS
40 PUBLICATIONS 244 CITATIONS
SEE PROFILE
SEE PROFILE
Paul Krueger
Rahim Moineddin
University of Toronto
University of Toronto
93 PUBLICATIONS 2,530 CITATIONS
457 PUBLICATIONS 10,075 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Pilot study on 18 F-FDG PET/CT for detection of inflammatory changes in blood-induced knee arthropathy in a rabbit model View
project
IT and healthcare View project
All content following this page was uploaded by David Kaplan on 06 June 2014.
The user has requested enhancement of the downloaded file.
Informatics in Primary Care 2011;19:241–50
# 2011 PHCSG, British Computer Society
Refereed paper
Using a data entry clerk to improve data
quality in primary care electronic medical
records: a pilot study
Michelle Greiver MD MSc CCFP
Director, North Toronto Research Network, Toronto, Canada and Assistant Professor, Department of
Family and Community Medicine, University of Toronto, Canada
Jan Barnsley BSc MES PhD
Associate Professor, Department of Health Policy, Management & Evaluation, University of Toronto,
Canada and Adjunct Scientist, Institute for Clinical Evaluative Sciences, Toronto, Canada
Babak Aliarzadeh MD MPH
Data Manager, North Toronto Research Network, Toronto, Canada
Paul Krueger BSc MHSc MSc PhD
Associate Professor, Associate Director Research Program, Department of Family and Community
Medicine, University of Toronto, Canada
Rahim Moineddin MSc PhD
Associate Professor, Department of Family and Community Medicine, University of Toronto, Canada,
Scientist, Institute for Clinical Evaluative Sciences, Toronto, Canada and Associate Professor, Dalla Lana
School of Public Health, University of Toronto, Canada
Debra A Butt MD MSc CCFP FCFP
Assistant Professor, Department of Family and Community Medicine, University of Toronto, Canada
Edita Dolabchian MSc
Research Associate, North Toronto Research Network, Toronto, Canada
Liisa Jaakkimainen MD MSc
Scientist, Institute for Clinical Evaluative Sciences, Institute for Clinical Evaluative Sciences, Toronto,
Canada and Associate Professor, Department of Family and Community Medicine, University of Toronto,
Canada
Karim Keshavjee MD MBA CCFP CPHIMS
Adjunct Assistant Professor, University of Victoria, Victoria, Canada
David White MD CCFP
Chief, Department of Family & Community Medicine, North York General Hospital, North York, Ontario,
Canada and Associate Professor, Family and Community Medicine, University of Toronto, Canada
David Kaplan MSc MD CCFP
Associate Family Physician-in-Chief, Chair of the Research Ethics Board, North York General Hospital,
North York, Ontraio, Canada and Assistant Professor, Department of Family and Community Medicine,
University of Toronto, Canada
For the North Toronto Research Network (NorTReN)
242
M Greiver, J Barnsley, B Aliarzadeh et al
ABSTRACT
Background The quality of electronic medical record (EMR) data is known to be problematic; research on improving these data is needed.
Objective The primary objective was to explore the
impact of using a data entry clerk to improve data
quality in primary care EMRs. The secondary objective was to evaluate the feasibility of implementing this intervention.
Methods We used a before and after design for this
pilot study. The participants were 13 community
based family physicians and four allied health
professionals in Toronto, Canada. Using queries
programmed by a data manager, a data clerk was
tasked with re-entering EMR information as coded
or structured data for chronic obstructive pulmonary
disease (COPD), smoking, specialist designations
and interprofessional encounter headers. We measured data quality before and three to six months
after the intervention. We evaluated feasibility by
measuring acceptability to clinicians and workload
for the clerk.
Results After the intervention, coded COPD entries
increased by 38% (P = 0.0001, 95% CI 23 to 51%);
identifiable data on smoking categories increased
by 27% (P = 0.0001, 95% CI 26 to 29%); referrals
with specialist designations increased by 20% (P =
0.0001, 95% CI 16 to 22%); and identifiable interprofessional headers increased by 10% (P = 0.45, 95
CI –3 to 23%). Overall, the intervention was rated as
being at least moderately useful and moderately
usable. The data entry clerk spent 127 hours restructuring data for 11 729 patients.
Conclusions Utilising a data manager for queries
and a data clerk to re-enter data led to improvements in EMR data quality. Clinicians found this
approach to be acceptable.
Keywords: computerised/standards, data collection/standards, data quality, health care/methods,
medical records systems, primary care, quality assurance
What is known
. Electronic medical record (EMR) data quality is known to be problematic.
. There are few interventional studies addressing this problem; interventions have generally led to modest
improvements.
What this paper adds
. Data queries programmed by a data manager followed by EMR data entry by a data clerk led to large
increases in structured data over a short period for a chronic disease (chronic obstructive pulmonary
disease), identifiable smoking categories and specialist designations.
. There was no significant increase in interprofessional encounter designations, a change that relied on
modification of clinician behaviour.
. The acceptability of the intervention to clinicians and the cost indicate that larger studies of similar
interventions are feasible.
Introduction
Data quality matters: ‘you cannot improve what you
cannot measure’.1 The transition from paper records
to electronic medical records (EMRs) has led to
expectations that electronic healthcare data collected
as part of routine practice will be available for quality
improvement activities, surveillance, research and
chronic disease management.2–12
The quality of the information collected fundamentally depends on the quality and integrity of data entered
in the charts. However, problems with the data include inconsistent or missing diagnostic coding and
risk factor designation, ‘dirty data’ (misspelled words,
inconsistent word strings, free text strings instead of
structured data), missing ‘meta-data’ (referral to ‘Dr
Smith’, where physician specialty is not listed) and
data entered in inconsistent or incorrect database
fields.13–21 During the transition to EMRs, training
is often focused on using and entering data in individual patient records, with limited emphasis on
consistent data entry and future auditing capabilities.
Family physicians and their practice teams may not be
aware of the importance of this issue22 and have many
competing demands on their time and resources.23
Once the EMR transition is complete, physicians may
have limited time, incentives or tools to modify and
Improving data quality in primary care EMRs
improve data that were initially entered as unstructured
free text, added to fields not meant for these specific
data or entered in several different areas of the EMR.
Systematic reviews of data quality have noted many
descriptive studies but few interventional studies designed to improve data quality in primary care EMRs.24,25
Most interventional studies used education or individualised feedback.25
Based on the existing literature and our clinical
experience, the underlying ideas for this study were:
(1) data entry difficulties were common during the
transition to EMRs; (2) problems were subsequently
not systematically corrected or managed; (3) a data
manager may be able to identify some problematic
areas; (4) trained data entry clerks could efficiently reenter data; and (5) once the initial data entry is done,
practices may be able to maintain reasonable data
quality using tools such as data manuals.
The primary purpose of this study was to explore
the impact of an intervention designed to improve
data quality in the EMRs of community based family
physicians. The secondary purpose was to evaluate the
feasibility of implementing this intervention.
Methods
Study design
We used a before and after design. We first used
professionally programmed data queries to measure
data quality and identify gaps. The intervention consisted of assigning data re-entry away from healthcare
providers: we used a data entry clerk for this work. We
then re-used the original queries after the intervention
so that the change could be calculated.
Participants
We recruited community based family physicians in
Toronto, Canada who were members of an interdisciplinary primary care organisation (the North
York Family Health Team) and were using the Nightingale On Demand1 EMR. Forty-three family physicians in the family health team used this software. We
recruited a convenience sample of 13 physicians that
have used EMRs for at least two years (to ensure that
early transition efforts were completed), as indicated
by the presence of EMR-based progress notes for over
two years. We also recruited four allied health providers who had provided clinical services to patients
registered to participating physicians during the study.
Eligible patients included all active patients registered
with the practices who were age 18 or more at the time
of the audit.
243
Intervention
The data queries were programmed by the North
Toronto Research Network (NorTReN) data manager; NorTReN is one of 10 practice-based research
networks currently participating in the Canadian Primary Care Sentinel Surveillance Network (CPCSSN),
Canada’s first multidisease primary care electronic
record surveillance system. A local data manager oversees EMR data collection, cleaning and transmission
to the central data repository for each network.13
We examined the change in data quality in four
areas of the EMR: diagnostic coding for a chronic
health condition (chronic obstructive pulmonary disease or COPD), structured categories for a risk factor
(smoking), structured specialist referral designation
(meta-data) and interprofessional encounter designation. The rationale for selecting these four areas is
that CPCSSN data managers have found that health
conditions are not consistently coded in the patient
health profile, smoking status is recorded using a large
number of free text terms, and specialist referral
designations are not consistently available.13 Interdisciplinary care provision is not currently collected
for CPCSSN, but is important for primary care system
planning.
In order to improve the generalisability of the findings,
we used data queries and extraction tools available
within the practices through their EMR interface. That
is, the data manager did not use queries that required
direct access to the underlying EMR databases, as this
method would not be available to practices wishing
to repeatedly query their own EMR for data quality
improvement purposes.
A research associate used the programmed queries
to audit the EMR and to record baseline measures. A
data clerk was then tasked with re-entering the data,
as follows: (1) with physician permission, adding the
International Classification of Diseases ninth revision
(ICD9) code 496 for COPD (the most common code
for this condition in the CPCSSN database) in the
patient health profile when free text indicating COPD
was found; (2) duplicating free text smoking data using
a drop down list classifying a patient as a current
smoker, ex-smoker or never smoked; (3) adding referral
designations to all specialists in the master referral list
that comply with College of Physicians and Surgeons
of Ontario specialist designation; and (4) adding standardised interprofessional encounter headers to the
EMR as a drop down list if these were not previously
present and informing allied health professionals. The
clerk was trained by the research associate and was
given data manuals with screen shots. The clerk entered
ten training records for each of the four areas which
were audited by the research associate for accuracy.
After the initial audits, clinicians were given data
manuals with suggested methods of data entry (see
244
M Greiver, J Barnsley, B Aliarzadeh et al
Appendix A available online at: www.radcliffepublishing.
com/journals/J12_Informatics_in_primary_care/
supplementary%20papers.htm).
Outcome measures
Data quality measures were: (1) COPD designations
in the patient health profile that were coded using
ICD9; (2) patient records that had data on tobacco use
in a structured format; (3) specialty referrals within a
three-month period with structured specialist designation; and (4) encounters by allied health professionals indicated as interprofessional care within a
three-month period.
We extracted EMR data at baseline and at three to
six months after the intervention.
We evaluated feasibility through acceptability to
clinicians and by measuring time and cost for the data
clerk. Clinicians (family physicians and allied health
providers) were given questionnaires incorporating
usefulness (perception of the degree that the process
would enhance job performance) and usability (perception of the degree that the process would be free from
effort).26,27 The questionnaires are shown in Appendix B available online at: www.radcliffepublishing.
com/journals/J12_Informatics_in_primary_care/
supplementary%20papers.htm. The clerk submitted
hours worked to the research associate; we recorded
total amount of time (including training) for each
data aspect.
Analysis
The significance of the change in the proportion of each
measure of quality was assessed using McNemar’s test
for paired samples. We used descriptive and summary
statistics for physician and practice characteristics and
for acceptability to clinicians. All tests were two-sided
and P-values < 0.05 were considered statistically significant. Data were analysed using SAS 9.2 (SAS Institute). This study was approved by the North York General Hospital’s Research Ethics Board. All physicians and allied health professionals who participated in the study provided signed, informed consent. Results Physician and practice characteristics Physician and practice characteristics are shown in Table 1. All physicians were in group practices. There were three office locations but physicians shared a single EMR server, accessed from a remote location. There were 11 729 eligible patients at the time of the initial audit in September 2010, and 11 554 patients at the time of the second audit in March 2011. A summary of the changes in data structure before and after the intervention is shown in Table 2. The Table 1 Physician and practice characteristics* Physician characteristics N=13 Female N (%) 10 (77) Age Median (range) 36 (34–59) CCFP N (%) 12 (92) Years since graduation from medical school Median (range) 11 (6–34) Canadian medical school graduate/foreign medical school graduate N/n 12/1 Number of physicians at the practice location Median (range) 5 (5–7) Number of nurses at the practice location Median (range) 1.4 (0.7–2.0) Duration of EMR use Median (range) 4 (2–7) Number of patients registered to the physician Median (range) 800 (660–1388) Number of patients seen in an average week Median (range) 80 (48–120) Number of hours providing office-based patient care per week Median (range) 25 (15–45) Note: CCFP = Certificate of the College of Family Physicians of Canada. * Obtained from self reports at study entry; based on full time equivalent. Improving data quality in primary care EMRs 245 Table 2 Coded or structured data present in the EMR Data element Baseline (%) Post intervention Difference*: % (%) (P, 95% CI) Coded COPD entries: number coded/total number with COPD in heath profile (%) 44/75 (59) 102/106 (96) 38 (P=0.0001, 23–51) Structured smoking categories: number with structured data/total number with smoking data (%) 4,285/6039 (71) 6208/6317 (98) 27 (P=0.0001, 26–29) Specialist designations in referral letters: number of structured designations/total number of specialist referrals (%) 831/1619 (51) 1177/1649 (71) 20 (P=0.0001, 16–22) 42/111 (38) 10 (P=0.45, –3–23) Interprofessional encounter headers: 25/89 (28) number of audited charts with appropriate headers/total number of audited charts (%) Note: CI = confidence interval; COPD = chronic obstructive pulmonary disease. * Difference may not be exact due to rounding. In a three month period. A 10% sample of interprofessional encounters for each of four allied health providers was randomly audited. proportion of coded or structured data elements increased for all categories studied, although this was not statistically significant for interprofessional encounter headers. Coded COPD entries in the patient health profile Prior to the intervention, 59% of COPD entries in the patient health profile were numerically coded using ICD9. ICD9 codes for this disease at baseline were 496, 492 and 491. The clerk entered all new codes as ICD9 496. The total number of COPD patients increased during the project because physicians concurrently verified patients with COPD as part of a quality assurance project in which they were given a list of patients who were potential COPD candidates because they were aged 45 years or over, non-asthmatic and used medications indicated for COPD (tiotropium, salbutamol, inhaled steroids).28 Those verified by physicians as having COPD were entered and coded by the clerk as ICD9 496. Data were re-audited in March 2011. The percentage of coded COPD entries increased to 96%. Pick list data on tobacco risk category Data about tobacco use were audited in October 2010. After the audit, the clerk accessed charts where free text tobacco information had been entered in the patient health profile and added data using a structured drop down list (current smoker, ex-smoker, never smoked). A follow-up audit occurred in March 2011. At baseline, 51% of patients had information on tobacco usage in their health profile, compared with 55% of patients during the second audit. Of those with tobacco data present, 71% had structured data on smoking; this was usually a checkbox indicating either smoker or non-smoker. After the intervention, 98% of patients with data on smoking had identifiable categories. Current smokers were not identifiable using standard EMR queries prior to the intervention. After the intervention, 732 patients (12% of those with data on smoking status) could be identified as current smokers. Structured specialist referral designations We audited the charts for a three-month period prior to the intervention (27 June 2010 to 27 September 2010). The data clerk added specialist designations to the master referral list in October 2010. We re-audited the charts for a three-month period following the intervention (15 November 2010 to 15 February 2011). One physician went on maternity leave between the first and second audits, and her data were censored from both audits. Identifiable specialist designations increased from 51 to 71%. 246 M Greiver, J Barnsley, B Aliarzadeh et al Standardised interprofessional encounter designations It was not possible to use the EMR software to audit encounters for the presence of interprofessional headers. EMR logs were used to identify all encounters done by an allied health provider for the three-month period prior to the intervention (27 June 2010 to 27 September 2010) and for a three-month period after the intervention (15 November 2010 to 15 February 2011). One allied health provider from each of the four categories in the family health team was randomly chosen, and then a randomly chosen 10% sample of the health provider’s encounters was manually audited. Usage varied by allied health provider role; the nurse almost never used the headers (2% of encounters prior to the intervention, 0% after), whereas the dietitian started using headers routinely (from 29% of encounters to 90% of encounters). The social worker and clinical pharmacist were using headers for all encounters prior to the intervention, and this did not change. Interprofessional headers increased by 10%; the change was not statistically significant. There were three questions dealing with usability and four questions dealing with usefulness. Data entry clerk workload Including training, the data entry clerk spent 3 hours recoding COPD, 53 hours restructuring the smoking data and 70 hours adding specialist designations to the master list. Interprofessional headers were added in less than 1 hour, for a total of 127 hours spent on all activities at a cost of $1905, or $147 per physician. COPD coding and tobacco categories required chart by chart data entry. The master referral list was shared between all practices as part of the common server, and therefore data were entered once for all practices. Interprofessional encounter headers were shared between all providers within an office but not between offices, and were therefore replicated three times for the three office locations studied. Discussion Participant ratings of interventions Participating clinicians rated the usability and usefulness of this approach. Results are shown in Table 3 for usability and Table 4 for usefulness. Eleven of the 13 eligible physicians returned the questionnaires: one physician did not respond and one was on maternity leave. All four allied health professionals responded. Principal findings A data manager can program queries to discover data quality issues and a trained data entry clerk can rapidly re-enter uncoded and unstructured data in the EMR as coded, structured and consistent data for groups of practices. We found significant increases in coded or Table 3 Participant rating of usability of intervention* Category {{ Rating: number of responses (% of all responses{ for each category ) Not at all usable Not very usable Neutral Moderately usable Very usable COPD coding 0 (0) 0 (0) 0 (0) 16 (48) 17 (52) Smoking category restructuring 0 (0) 0 (0) 1 (3) 8 (24) 24 (73) Specialist designation 0 (0) 0 (0) 4 (12) 14 (42) 15 (45) Interprofessional 0 (0) encounter headers 0 (0) 0 (0) 3 (25) 9 (75) * Obtained from self reported perceptions of usability {{and usefulness28 at study exit. { Percentage may not add up to 100 due to rounding. From 11 physicians and 4 allied health professionals; there were three questions for each category, for a total of 33 responses from physicians and 12 responses from allied health professionals. Improving data quality in primary care EMRs 247 Table 4: Participant rating of usefulness of intervention* Category {{ Rating: number of responses (% of all responses{ for each category ) Not at all useful Not very useful Neutral Moderately useful Very useful COPD coding 0 (0) 4 (9) 4 (9) 9 (20) 27 (61) Smoking category restructuring 0 (0) 4 (9) 9 (20) 7 (16) 24 (55) Specialist designation 0 (0) 2 (5) 16 (36) 13 (30) 12 (27) Interprofessional 2 (13) encounter headers 0 (0) 3 (19) 9 (56) 2 (13) * Obtained from self reported perceptions of usability {{and usefulness28 at study exit. { Percentage may not add up to 100 due to rounding. From 11 physicians and 4 allied health professionals; there were four questions for each category, for a total of 44 responses from physicians and 16 responses for allied health professionals. structured data with this intervention for three of the four areas we studied. There was good acceptance by providers, and the time spent by the data clerk did not seem excessive. These findings indicate that a larger study would be feasible. Implications of the findings Uncoded data were present in all four areas we studied for these community based family physicians. Standardising data elements may assist in developing comparisons over time within practices, between individual providers, between groups of physicians and between different jurisdictions.21 An unexpected finding was that a list of current smokers could not be generated using EMR queries prior to the intervention, due to the variability in data entry. This was possible after data entry. The ability to identify a population at risk can enable EMR features associated with improved quality of care such as chart alerts or recalls.29–31 We believe that the magnitude of the changes is likely clinically important. A change of 5% or more has been used to determine the minimal clinically important difference32–34 and we report larger changes in this study. Comparison with the literature Difficulties with coding have been reported previously.14,16,19,20,35 Other studies have found that the use of data extraction was possible; similar to our study, practices required external support for this.36 Studies on data improvement in primary care EMRs have relied on audit, feedback and training, with moderate effects.37–41 In this study, we report larger changes through the involvement of non-clinicians in data management and data re-entry in EMRs. The area that required a change in clinician data entry behaviour (interprofessional encounter headers) did not change significantly. Limitations Limitations for this pilot study include lack of population diversity and a single EMR system. However, the fact that the study took place in a community based primary care setting indicates the potential to conduct a larger community based trial that would be more broadly generalisable. There was a concurrent quality assurance effort for COPD. This increased the total number of patients defined as having this condition (denominator) after the intervention and may have affected the results for this aspect of data quality. The information is limited to the EMR application we studied. However, we believe that similar restructuring can be undertaken with other EMR software applications commonly used in primary care; data issues have been found with every EMR studied as part of CPCSSN.13,19 The cost and workload of the data management and data entry clerk, as well as the amount of training 248 M Greiver, J Barnsley, B Aliarzadeh et al required will vary according to the EMR type, the structure of the EMR used, provider data entry habits and availability of technical resources. In this study, we did not validate the accuracy of data or address data completeness. In Toronto, 18.2% of persons age 12 years or over are current smokers42 compared to 12% of adults with smoking data identified in our study, so it is possible that there was missing or misidentified data on current smokers in the EMR. We tested the overall effect of the implementation of our intervention. We did not capture impact on provider behaviour, such as changes in data entry habits, for three of the four measures. Members of primary care practices continually enter data, and may reverse some of the improvements over time. We measured data over a relatively brief interval and longer studies will be needed to quantify the loss of data quality. However, the physician’s generally positive ratings of usefulness suggest that improvements in data quality provide a recognisable benefit to EMR users, who may thus improve their own data entry. Interventions such as repeated audit and feedback, as well as ongoing maintenance activities (such as the use of data clerks at fixed intervals) would be needed to assess and maintain data quality. Further research and recommendations Additional studies and methods could include measures of data entry reliability such as re-audits of data entry samples; measures of validity (comparison with reference standard); sustainability of the changes; qualitative methods to explore perceptions and barriers to this approach; impact on provider behaviour such as improved quality and consistency of data entry; and an economic analysis of the cost of data entry clerks in various primary care settings and EMR applications. Collaboration with EMR vendors to improve the structure of their underlying databases may be worthwhile. Efforts are underway to implement EMR data content standards.43,44 Vendors could increase the amount of structured data that can be captured as part of clinical care in their applications, could improve queries and could automate data linkages. For example, allied health professionals could automatically have searchable designations linked to encounters. Conclusions In this study, the use of a data entry clerk led to fairly large and rapid improvements in EMR data quality. We found increases in COPD coding, standardised tobacco risk categories and structured specialist designation, with reasonable rates of clinician acceptance and workload for the clerk. ACKNOWLEDGEMENTS This study was funded by the Primary Health Care System Program; we acknowledge the support of the Ontario Ministry of Health and Long-Term Care. The views expressed do not necessarily reflect those of the Ministry. The Canadian Primary Care Sentinel Surveillance Network is funded by the Public Health Agency of Canada. The views expressed herein do not necessarily represent the views of the Public Health Agency of Canada. REFERENCES 1 Newcomer LN. Perspective: physician, measure thyself. Health Affairs 1998;17(4):32–5. 2 Romanow R. Building on Values: the future of health care in Canada. Ottawa: National Library of Canada, 2002. 3 Garcia Rodriguez LA and Perez Gutthann S. Use of the UK General Practice Research Database for pharmacoepidemiology. British Journal of Clinical Pharmacology 1998;45(5):419–25. 4 Bartholomeeusen S, Kim CY, Mertens R, Faes C and Buntinx F. The denominator in general practice, a new approach from the Intego database. Family Practice 2005;22(4):442–7. 5 Biermans MC, Spreeuwenberg P, Verheij RA, de Bakker DH, de Vries Robbe PF and Zielhuis GA. Striking trends in the incidence of health problems in The Netherlands (2002–05). Findings from a new strategy for surveillance in general practice. European Journal of Public Health 2009;19(3):290–6. 6 Hueston WJ, Mainous III AG, Ornstein S, Pan Q and Jenkins R. Antibiotics for upper respiratory tract infections: follow-up utilization and antibiotic use. Archives of Family Medicine 1999;8(5):426–30. 7 Hall J, Tomlin A, Martin I and Tilyard M. A general practice minimum data set for New Zealand. New Zealand Medical Journal 2002;115(1163):U200. 8 Stocks NP, McElroy H, Sayer GP and Duszynski K. Acute bronchitis in Australian general practice. A prescription too far? Australian Family Physician 2004; 33(1–2):91–3. 9 Lagasse R, Desmet M, Jamoulle M et al. European Situation of Routine Medical Data Collection and their Utilization for Health Monitoring: Euro-Med-Data Final Report. Universite Libre de Bruxelles: Brussels, 2001. 10 Smeeth L, Thomas SL, Hall AJ, Hubbard R, Farrington P and Vallance P. Risk of myocardial infarction and stroke after acute infection or vaccination. New England Journal of Medicine 2004;351(25):2611–18. 11 Schade R, Andersohn F, Suissa S, Haverkamp W and Garbe E. Dopamine agonists and the risk of cardiacvalve regurgitation. New England Journal of Medicine 2007;356(1):29–38. Improving data quality in primary care EMRs 12 Have paper records passed their expiry date? Canadian Medical Association Journal 2005;173(7):725. 13 Birtwhistle R, Keshavjee K, Lambert-Lanning A et al. Building a pan-Canadian primary care sentinel surveillance network: initial development and moving forward. Journal of the American Board of Family Medicine 2009; 22(4):412–22. 14 Jordan K, Porcheret M and Croft P. Quality of morbidity coding in general practice computerized medical records: a systematic review. Family Practice 2004;21(4): 396–412. 15 Whitelaw FG, Nevin SL, Milne RM, Taylor RJ, Taylor MW and Watt AH. Completeness and accuracy of morbidity and repeat prescribing records held on general practice computers in Scotland. British Journal of General Practice 1996;46(404):181–6. 16 Selak V, Wells S, Whittaker R and Stewart A. Smoking status recording in GP electronic records: the unrealised potential. Informatics in Primary Care 2006;14(4):235– 41; discussion 242. 17 Mant J, Murphy M, Rose P and Vessey M: The accuracy of general practitioner records of smoking and alcohol use: comparison with patient questionnaires. Journal of Public Health Medicine 2000;22(2):198–201. 18 Hollowell J. The General Practice Research Database: quality of morbidity data. Population Trends 1997(87): 36–40. 19 Mukhi S and Keshavjee K. Developing a Primary Care Electronic Medical Record Chronic Disease Surveillance System in Canada: data quality and lessons learned. Canadian Association of Health Services and Policy Research: Calgary, 2009. 20 Birtwhistle R, Keshavjee K, Martin K and LambertLanning A. Improving Data Quality in EMRs for Chronic Disease Management. Family Medicine Forum: Calgary, 2009. 21 Canadian Institutes for Health Information. CPCSSN Data Analysis Report, Cycle 2A. Canadian Institutes for Health Information: Toronto, 2010. 22 Audet AM, Doty MM, Shamasdin J and Schoenbaum SC. Measure, learn, and improve: physicians’ involvement in quality improvement. Health Affairs (Millwood) 2005;24(3):843–53. 23 Yarnall KS, Pollak KI, Ostbye T, Krause KM and Michener JL. Primary care: is there enough time for prevention? American Journal of Public Health 2003; 93(4):635–41. 24 Thiru K, Hassey A and Sullivan F. Systematic review of scope and quality of electronic patient record data in primary care. BMJ 2003;326(7398):1070. 25 Brouwer H, Bindels P and Weert H. Data quality improvement in general practice. Family Practice 2006; 23(5):529–36. 26 Bagozzi RP, Davis FD and Warshaw PR. Development and test of a theory of technological learning and usage. Human Relations 1992;45(7):659–86. 27 Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly 1989;13(3):319–39. 28 Qaseem A, Wilt TJ, Weinberger SE et al. Diagnosis and management of stable chronic obstructive pulmonary disease: a clinical practice guideline update from the 249 American College of Physicians, American College of Chest Physicians, American Thoracic Society, and European Respiratory Society. Annals of Internal Medicine 2011;155(3):179–91. 29 Szilagyi PG, Bordley C, Vann JC et al. Effect of patient reminder/recall interventions on immunization rates: a review. Journal of the American Medical Association 2000; 284(14):1820–7. 30 Kaczorowski J, Goldberg O and Mai V. Pay-for-performance incentives for preventive care. Canadian Family Physician 2011;57(6):690–6. 31 Holbrook A, Thabane L, Keshavjee K et al. Investigators ftCI: individualized electronic decision support and reminders to improve diabetes care in the community: COMPETE II randomized trial. Canadian Medical Association Journal 2009;181(1–2):37–44. 32 Copay AG, Subach BR, Glassman SD, Polly DW Jr and Schuler TC. Understanding the minimum clinically important difference: a review of concepts and methods. Spine Journal 2007;7(5):541–6. 33 Jaeschke R, Singer J and Guyatt GH. Measurement of health status: ascertaining the minimal clinically important difference. Controlled Clinical Trials 1989;10(4): 407–15. 34 Greiver M, Kasperski J, Barnsley J and Rachlis V. Measuring Quality Improvements in Preventive Care Services in the First Two Family Health Networks in the Greater Toronto Area. Ontario College of Family Physicians: Toronto, 2006. 35 Khan NF, Harrison SE and Rose PW. Validity of diagnostic coding within the General Practice Research Database: a systematic review. British Journal of General Practice 2010;60(572):e128–36. 36 Schattner P, Saunders M, Stanger L, Speak M and Russo K. Clinical data extraction and feedback in general practice: a case study from Australian primary care. Informatics in Primary Care 2011;18(3):205–12. 37 Linder JA, Rigotti NA, Schneider LI et al. An electronic health record-based intervention to improve tobacco treatment in primary care: a cluster-randomized controlled trial. Archives of Internal Medicine 2009;169(8): 781–7. 38 de Lusignan S, Stephens PN, Adal N and Majeed A. Does feedback improve the quality of computerized medical records in primary care? Journal of the American Medical Informatics Association 2002;9(4):395–401. 39 Porcheret M, Hughes R, Evans D et al and the North Staffordshire General Practice Research Network. Data quality of general practice electronic health records: the impact of a program of assessments, feedback, and training. Journal of the American Medical Informatics Association 2004; 11(1):78–86. 40 Lopez-Picazo Ferrer JJ, Agullo Roca F, Villaescusa Pedemonte M and Cerezo Corbalan JM. Clinical data that are essential for the primary care clinical records: an experience of evaluation and improvement. Aten Primaria 2002;30(2):92–8. 41 Brown PJ and Warmington V. Info-tsunami: surviving the storm with data quality probes. Informatics in Primary Care 2003;11(4):229–33; discussion 234. 42 Table 105–0502 – Health indicator profile, two year period estimates, by age group and sex, Canada, prov- 250 M Greiver, J Barnsley, B Aliarzadeh et al inces, territories, health regions (2007 boundaries) and peer groups, occasional. Ottawa: Statistics Canada, 2010. 43 Sullivan-Taylor P, Flanagan T, Harrison T and Webster G. Development of a draft pan-Canadian primary health care electronic medical record content standard. Studies in Health Technology and Informatics 2011;164:385–91. 44 Sullivan-Taylor P, Mukhi S, Martin-Rhee M and Webster G. Data that makes a difference in quality improvements in primary health care: approaches through a pan-Canadian voluntary electronic medical record source. Studies in Health Technology and Informatics 2011;164:367–71. ADDRESS FOR CORRESPONDENCE Michelle Greiver North Toronto Research Network 240 Duncan Mill Road Suite 705 Toronto Ontario M3B 3S6 Canada Email: mgreiver@rogers.com Accepted January 2012 Supplementary material is available online at: www.radcliffepublishing.com/journals/J12_Informatics_ in_primary_care/supplementary%20papers.htm View publication stats

Essay writing service:

• Excellent quality
• 100% Turnitin-safe
• Affordable prices

Order Now