Statistical Thinking in Health Care
Business Statistic MAT510
February 1, 2015

HMO’s pharmacy is experiencing problems with providing inaccurate prescriptions to customers. The inaccurate prescriptions could be from a number of areas in the prescription process. The errors could stem from interpretation of the prescribed medication to faulty input into the computer system, or even the incorrect understanding of the medication prescribed. To develop strategies to target the prevention of medication errors fully, it is necessary to have a holistic understanding of the medication-use process in the pharmacy and how each stage contributes to the overall error rate. Drug ordering and delivery are typically broken into four different stages; prescribing, transcribing, filling, and dispensing. Each of these stages represents a possible risk point and a potential vulnerable link in the patient-safety chain. The below process map outlining the steps HMO takes to fill a prescription for the customer.
Prescribed
Received
Processed
Filled
Dispensed

Suppliers | Inputs | Process | Outputs | Customers | Doctor | Prescription | Medication prescribed by doctor | Medication | Patients | Pharmaceutical Company’s | Drugs | Receiving prescription | Payment | | | | Interpret | | | | | Transferring into computer | | | | | Processing against insurance | | | | | Filling medication | | | | | Dispensing Medication | | |

The analysis of the charts included allows for an understanding of the pharmacy process; but it just might not be the fault of HMO pharmacy. Many medication errors result from prescribing errors, which have an increased potential for serious complications. Prescribing errors are classified into different categories based on knowledge, rules, action, and memory (Warholak, Queiruga, Roush, & Phan, 2011). Knowledge-based errors reflect lack of experience or understanding about certain medications. Rule-based errors reflect lack of application of fundamental rules. Action based errors are those that are not intended a misspelling or mistaken drug name. Memory-based errors involve forgotten information; for example patient allergy to medication. The doctor prescribes the medication and the customer brings the hand script to HMO pharmacy for processing. Once HMO pharmacy receives the prescription the script is loaded into the computer system by the pharmacy assistant and then it is processed against the customer’s insurance and availability of the medication. The pharmacist fills the medication and it is dispensed to the customer. A possible main root cause of the inaccurate prescriptions could stem from the hand written scripts by the doctor. The doctor’s illegible handwriting and unclear instructions has the potential to be the root cause of the problem. This causes the pharmacy assistant to input the medication into the computer system incorrectly; therefore the pharmacist processes and dispenses the wrong medication.

The horrible hand script error could be coming from one or many doctors. To categorize the illegible handwriting and unclear instructions a study would have to be done on all hand scripts processed at HMO pharmacy to see if there was a trend with all doctors or if it could be narrowed down to a select few doctors. If the data supports one doctor’s handwriting or even a group of doctors that are causing all the inaccurate prescriptions it would be termed as a common cause case (Hoerl & Snee, 2012). A common cause case variation is always present, has numerous sources, it is part of normal behavior in the process and the process is stable. If the data gathered could not be pin pointed to one doctor or a group of doctors but it was a random error occurring with no apparent pattern it would be a special cause variation. A special cause is normally temporary, unpredictable and unstable process. The study’s data indicated HMO’s problem is linked to numerous doctors with illegible handwriting and unclear instructions. Although each doctor’s handwriting was not always illegible; therefore this is a special cause variation. Since the data is unpredictable and unstable it is categorized as special cause.
The main method to gather data would be an observational study of hand written prescriptions processed at HMO pharmacy (Reisman, Gienapp, & Stachowiak, 2012). The study would identify the doctor who wrote the script, the medication prescribed, the legibility of the handwriting, and the instructions for the medication. The script would then be verified for correctness based on data gathered. The study would include scripts from all area doctors over a month. Analysis of the data would determine if the error scripts were clustered by a group of doctors, a specific office or hospital, or even a single doctor. The study would also show if the errors were random and not tied to a specific group, office/hospital, or one individual doctor. Additionally, a questionnaire would be developed and sent to other local pharmacies to identify if the data gathered was a systemic problem or just with HMO’s interpretation of the handwritten prescriptions.
The solution to fix HMO pharmacy problem of inaccurate prescriptions is to institute an electronic means of transferring prescriptions from the doctor’s office, hospital, or clinic to the pharmacy (Bubalo, et al., 2014). This would eliminate handwritten scripts and unclear instructions. The doctors would be able to submit the scripts in the computer system directly to the pharmacy. If the medication was not available, or not in the dose prescribed, or instructions were not clear, the pharmacist or the pharmacy assistant would be able to contract the doctor via the same electronic means.
The strategy proposed to measure this solution would be to document any time the pharmacy assistant or the pharmacist went back to the doctor for clarification or to ask questions. Another way to measure would be how many incorrect prescriptions were reported with the new system vice how HMO was processing mediation over a six month timeframe. These two strategies would allow for HMO to correctly gage the effectiveness of the solution.

References

Bubalo, J., Warden, B. A., Nishida, T., Nguyen, L., Edillo, N., Wiegel, J. J., & Svoboda, L. M. (2014). Does Applying Technology Throughout the Medication use Process Improve Patient Safety with Antineoplastics. Oncology Pharmacy Practice, 445-460.
Hoerl, R., & Snee, R. (2012). Statistical Thinking . Hoboken: John Wiley and Sons.
Reisman, J., Gienapp, A., & Stachowiak, S. (2012). A Handbook of Data Collection Tools. Retrieved from Annie E. Casey Foundation: http://www.organizationalresearch.com/publicationsandresources/a_handbook_of_data_collection_tools.pdf
Warholak, T. L., Queiruga, C., Roush, R., & Phan, H. (2011). Medication Error Identification Raes by Pharmacy, Medical, and Nursing Students. American Journal of Pharmaceutical Education, 75.