Automated Screening of Patient’s Electronic Medical Records Increases the Quality While Reducing The Cost Of Medical Testing Essay

Introduction

Aspden, Corrigan et al. (2004), in their 2003 Institute of Medicine of the National Academies patient safety report describes an electronic medical records as a system which encompasses, “a longitudinal collection of electronic health information for and about persons which can be immediately accessed by people with the help of an authorized user and can help in provision of knowledge and decision-support aimed at enhancing the quality, safety & efficiency of patient care and support for efficient processes for health care delivery”¹.

This paper is a grant proposal whose objective is to assess the effect of automated screening of patient’s electronic medical records on the quality of healthcare delivery and the cost of medical testing services.

Hypothesis

The study hypothesis is that automated screening of outpatient’s electronic medical records in various clinics in Midwestern USA increases the quality of care of the patients with chronic illnesses while reducing the cost of their medical testing by 40% within a period of one years.

Literature Review

The Institute of Medicine, Deloitte research and other research firms, have indicated that despite health care being an information intensive industry, it remains highly fragmented and inefficient in comparison to the other sectors such as banking, insurance and education, but nevertheless, major changes to reform the health care digital information systems have started to take centre stage².

These changes are aimed at rapidly rising health care costs and increasing concerns for patient’s safety and the quality of health care³. The use of information technology to address improvements in patient safety is currently driving health care organizations to automate clinical care operations and associated administrative functions⁴.

Research findings published in the MEDLINE database from the Journal of American Board of Family Practice by Wilbur, Huffman, Lofto and Finnell in the year 2011 in their study, found out that, “of the 8,489 emergency department patients who participated in the study, 5,794 (68.3%) of them were identified to be eligible for screening”⁵.

The research also found out that, “of the 1,484 (25.6%) patients approached for screening, 1,121 (75.5%) consented to be screened for HIV Virus and 5 (0.4%) received confirmed positive results”⁶.

The research also revealed that, “the reasons for ineligibility, as determined by the electronic medical record system, were previous screening 1,125 (41.7%), age 890 (33.0%), known HIV 111 (4.1%) and those with unknown reasons were 569 (21.1%)”⁷.

It was concluded that, “clinical informatics solutions can provide automated delineation of emergency departments subpopulations eligible for HIV screening, according to predetermined criteria, which could increase program efficiency and might accelerate integration of HIV screening into clinical practice”⁸.

Research done by Liljeqvist et al. in 2011, indicated that, “syndrome influenza-like illness data can be extracted automatically from routine general practitioners data and that how influenza-like illness trends in sentinel general practice compares with influenza-like illness trends in emergency departments has to be monitored using electronic medical records”.

This research indicated that, “the general practice surveillance tool identified seasonal trends in influenza-like illness both retrospectively and in real-time”⁹. The number of weekly influenza-like illness presentations ranged from 8 to 128 at general practice sites and from 0 to 18 in emergency departments in non-pandemic years¹⁰.

It was concluded that, “Automated data extraction from routine general practice records offers a means to gather data without introducing any additional work for the practitioner”¹¹. Therefore, adding this method to current surveillance programs will enhance their ability to monitor influenza-like illness to detect early warning signals of new influenza-like illness events in the population¹².

Ebbing, Woeltje & Malani, (2010) asserted that, “individual components of an automated screening system could identify Surgical Site Infection Surveillance”.

Sands et al. indicated that, “the use of coded diagnoses, tests, and treatments in the medication record had a sensitivity of about 74% and that specific codes and combinations of codes identified a subset of 2% of all procedures, among which 74% of Surgical Site Infection Surveillance occurred and that the use of hospital discharge diagnosis codes and pharmacy dispensing data had a sensitivity of 77% and a specificity of 94%¹³.

In another research conducted by Platt and his associates it was indicated that, “automated claims and pharmacy data from several health insurance plans can be combined to allow routine monitoring for indicators of postoperative infections” ¹⁴.

Buuam demonstrated that, “cooperation between infection control and medical informatics personnel can produce an automated system of surveillance which requires much less time and performed well with sensitivity and a specificity of 91%¹⁵.

In a research to find out if automation of patients electronic records improves surrogate patient’s outcomes in outpatient settings done by Anthony Jerant and David Hill, it was indicated that, “utilization of either complete or hybrid Electronic Medical Records can improve some surrogate outpatient’s care outcomes”¹⁶.

Research Methodology

In order to collect rational data, purposive sampling method will be employed. 250 diabetic terminally ill patients who will have attended clinic for at least a year will be used as a sample population for the study.

The respondents and the medical clinics to participate in the study will be randomly selected within Midwest region of USA¹⁷. Two physicians from each of the ten selected clinics will be selected as key respondents to give first hand information on how the automated screening of outpatient’s electronic medical records for the terminally ill patients affects the care and cost of medication of those patients¹⁸.

Data collection from the 250 diabetic terminally ill patients who will have been selected using purposive sampling methodology will be done by five research assistants who will each interview 50 patients.

I will be the project manager of the exercise and I will be involved in planning, organizing, staffing, directing, controlling, reporting and budgeting which are vital activities in making sure that the exercise is a success.

This exercise will run for 3 months. The first two weeks will be utilized to seek legal permission from research governing bodies. The next 5 weeks will be used to send structured questionnaire to the key respondents. Week 7 up to week 8 will be used to carry out interviews of the 250 diabetic terminally ill respondents. Week 9 to week 12 will be used for data analysis and reporting of the research findings.

Research Personnel

The principle investigator will be an experienced statistician with years of experience in data collection and analysis. His or her duty will be to be in charge of the research team and will supervise them and insure that the process is going according to the regulations governing social scientific research. He or she will be totally committed to the process from initiation to the reporting cycle of the project.

The Principle researcher will send the structured question­naires to the selected key respondents. The Principle researcher will train the research assistants for a period of 5 days and also supervise them during data collection for a period of 5 days. The principle researcher will also conduct data storage, analysis and reporting.

The research assistants who will work under the principal investigator will carry out interview of the 250 selected diabetic terminally ill respondents mentioned above. The research assistants will only be part of the process during the data collection phase and they will be trained by the principal investigators for 3 days on how to use data collection tools i.e. interview questionnaires.

The project will also engage the services of a clerical staff whose responsibility will be to keep an inventory of the logistics and costs associated with the project in order to insure transparence and accountability of expenditure.

Data Collection and Storage

Question­naires will be mailed to the 20 selected physician homes. Preaddressed, postage-paid return enve­lopes will be provided and follow-up mailing will be done to their offices 2 weeks later. These reminders will be conveyed via email, weekly newsletters and staff meetings. Information from the 20 key respondents will be expected after 5 weeks from the time the first mailing process is done¹⁹.

The ten selected medical clinics will be contacted and booked for the exercise. The clinic’s management will be requested to allow the research team to interview twenty five terminally ill patients from each of the ten selected clinics who meet the condition that, ‘the respondent must have attended the clinic for at least one year’²⁰.

Five research assistants will collect data from fifty respondents for at least ten hours a day for a period of five days.

The data will be stored on a personal computer and later on analyzed using SPSS Version 11.0.1 software²¹.

Data Analysis and Evaluation

Data will be analyzed appropriately based on the category of responses from the various respondents²². The data will later be triangulated with the medical clinics reports data. The findings of the study will form part of the final research study report. The findings of this study will be disseminated to the study respondents and also to medical data bases such as MEDLAB in order to be used in decision making and policy making²³.

References

1. Aspden, P., Corrigan, J., Wolcott, J. and Shari, M. (2004). Patient safety: Achieving a new standard for care, Committee on data standards for patient safety and board on health care services. Washington, DC: National Academy Press.

2. United States general Accounting office. (2003). Information technology benefits realized for selected health care functions. USA: Diane publishing company.

3. Wilbur, L., Huffman, G., Lofto, S. and Finnell, J. (2011). The use of a computer reminder system in an emergency department universal HIV screening program. Annals of emergency medicine, 58(1), 71-73.

4. Liljeqvist, G., Staff, M., Puech, M., Blom, H. and Torvaldsen, S. (2011). Automated data extraction from general practice records in an Australian setting: trends in influenza-like illness in sentinel general practices and emergency departments. BMC Public health journal, 6, 11:435.

5. Peterson, L., Hacek, D., Rolland, D. and Brossette, S. (2003). Detection of a community infection outbreak with surveillance. Lancet, 362(1), 1587-1588.

6. Goatham, J., Smith, F., Birkhead, S. and Daisson, C. (2003). Policy issues in developing information systems for public health surveillance of communicable diseases. New York: Springer publishers.

7. Jerant, A. and Hill, D. (2000). Does the Use of Electronic Medical Records Improve Surrogate Patient Outcomes in Outpatient Settings? The journal of family practice, 49(4), 349-357.

8. Immy, H. and Wheeler, S. (2010). Qualitative research in nursing and healthcare. Ames, Iowa: Wiley-Blackwell.

9. Teddlie, C. and Tashakkori, A. (2009). Foundations of mixed methods research: integrating quantitative and qualitative approaches in the social and behavioral sciences. Los Angeles: SAGE.

10. Vogt, P. (2010). Data collection. Los Angeles: SAGE.

11. Griffith, A. (2010). SPSS. Hoboken : John Wiley & Sons.

12. Lee, T. and Wang, J. (2003). Statistical methods for survival data analysis. Hoboken, N.J.: Wiley.

Footnotes

¹ (Aspden, Corrigan, Wolcott & Shari, 2004).

² (United States general Accounting office, 2003).

³ (United States general Accounting office, 2003).

⁴ (United States general Accounting office, 2003).

⁵ (Wilbur, L., Huffman, G., Lofto, S. & Finnell, J., 2011).

⁶ (Wilbur, L., Huffman, G., Lofto, S. & Finnell, J., 2011).

⁷ (Wilbur, L., Huffman, G., Lofto, S. & Finnell, J., 2011).

⁸ (Wilbur, L., Huffman, G., Lofto, S. & Finnell, J., 2011).

⁹ (Liljeqvist, Staff, Puech, Blom & Torvaldsen, 2011).

¹⁰ (Liljeqvist, Staff, Puech, Blom & Torvaldsen, 2011).

¹¹ (Liljeqvist, Staff, Puech, Blom & Torvaldsen, 2011).

¹² (Liljeqvist, Staff, Puech, Blom & Torvaldsen, 2011).

¹³ (Peterson, Hacek, Rolland & Brossette, 2003)

¹⁴ (Goatham, Smith, Birkhead, & Daisson, 2003)

¹⁵ (Goatham, Smith, Birkhead, & Daisson, 2003)

¹⁶ (Jerant & Hill, 2000).

¹⁷ (Immy, & Wheeler, 2010)

¹⁸ (Teddlie &Tashakkori, 2009).

¹⁹ (Vogt, 2010).

²⁰ (Vogt, 2010).

²¹ (Griffith, 2010).

²² (Lee & Wang, 2003).

²³ (Lee & Wang, 2003).