Variable Adjustment Life Display in Healthcare - Coursework Example

Analysis of VLAD Data Introduction Variable adjustment life display is a quality monitoring tool designed to show issues that need further review. It is a tool for screen applied to clinical indicators to show areas with high risks of the services and quality of the services. Its introduction in Queensland health had an objective in assisting monitoring the quality of services to the patients. The method can monitor one individual or a unit (Sorensen & Iedema, 2011). The method is vital in finding ways to improve the quality of health services. VLAD shows clinical outcomes in graphical technique that is easy to understand. The plot shows the cumulative differences in a variable between the actual and expected outcome. The technique cumulatively incorporates information on the risk of each case. The method also finds application in real-time monitoring in high-risk operations (Roberts, Tang, Harvey & Kadirkamanathan, 2012). The method involves flagging when variations reach the three pre-specified levels. However, the flags do not immediately indicate good or bad performance as there are other possible reasons for flagging. There is the corporative submission of patients’ data to contract VLAD usually from many hospitals usually through the database (Queensland Government, 2014). Update of these data is on the monthly basis that allows early detection of any area of concern and fast improvement of service quality. The early detection of variation in the outcome through VLAD is not usually there in the block audit. Investigation after flagging done by the hospital staff is mainly through the Pyramid Model of Investigation. The model has data at the bottom have the higher possibility of causes than the professional at the top. VLAD plays a vital role also in determining the difference between one hospital and the others through the use of control limits. In the surgical situation, it can track the outcome of surgeon over time (Browse, Black, Burnand, Corbett & Thomas, 2010). Interpretation of VLAD Each chart of VLAD shows presentation of specific indicator in the health system. The preparation of the chart uses the cumulative difference in the outcome of the patients and has upper and lower limits. These limits give an idea whether the performance in the hospital is better or worse (Foltran, Baldi, Bertolini, Merletti & Gregori, 2009). The limits will suggest the extent of the condition. The chart plotting is in a way that the clinical indicator is against the period for a particular hospital. Each patient outcome has upward or downward contribution to the graph line through the difference from the expected. The successful outcome causes the graph to rise while the unsuccessful one causes it to drop. The overall extents of the line movement depend on the cumulative difference between the expected and actual outcome. Calculation of the difference is through subtracting the observed outcome from the expected outcome (Faltin, Kenett & Ruggeri, 2012). In mortality indicator, the graph shows statistical estimates of lives gained. Generation of the expected outcome figure is on the basis of organization statistics and risk profile. Signalling take place when either red (lower control limit) or blue (upper control limit) line intercept the VLAD line (Andrianopoulos, Jolley, Evans, Brand & Cameron, 2012). The black line is VLAD that shows the estimated number of the cumulative difference of the deaths obtained in the hospital from a stroke. If these limits intersect with VLAD line, then it signals “out-of-control” state. Hence hospital needs to access quality of the services. The limit lines are also important in flagging when they touch VLAD line. The positioning of the flag at a certain point shows that patients get unexpected for a given period. The upper flags show that the hospital has patient outcome rate is lower than the stated outcome rate. The line showing downward slope in the graph shows worsening mortality experiences in the hospital (ITCH, McDaniel & IOS Press, 2009). A straight horizontal line in the graph shows no changes in the outcome from the previous ones. The flag on the lower side shows the hospital has a higher rate of outcome than stated. There are many indicator groups in VLAD for analysis (Healy, 2013). Stroke in-hospital mortality is in indicator group B where triggering of the flags occurs at the 30%, 50% and 75% variation from the average state (Sabin, 2008). There are usually four types of indicator in the VLAD that have differing flagging levels. Indicator group A has flag triggering level at 10%, 20% and 30%. Indicator Group B has flags triggering levels at 30%, 50% and 75%. Indicator group C has flags triggering levels at 50%, 75% and 100 percent. Indicator group D has flag triggering levels at 100%, 125% and 150%. Each level has a specific illness that they contain. Estimation of the predicted outcome is very important and needs to be accurate. Any exaggeration or inaccurate approximation of the expected outcome leads to error even in flagging. The flag at the point of interception marks points in which the prior VLAD need review. The case at the point where there is the interception of lines does not show that the case at that time is the cause of variation. In normal representation, flagging occurs as a result of cumulative effects of the outcome. The interception calls for review giving opportunity to explore the cause the positive or negative effect on stroke in-hospital mortality. The table shown above is stroke in-hospital VLAD. The y-axis has both positive and negative numbers in different directions. Increasing negative numbers show a rise in mortality levels in the hospital due to stroke. The y-axis shows an estimation of the lives gained. The x-axis shows the number of cases of stroke in the hospital. The rise or fall in the y-axis shows the situation direction. The slopes show changes in performance (Lawless, 2014). Rising slope shows the mortality rate is getting better than the previous time. In other types of VLAD graphs, rising slope shows improving outcome results. The graph does not provide exact proof of effect but show point for investigation (WHO, 2012). The VLAD graph above has two points of interceptions. One of the points is on the upper side while the other one is on the lower side. An interception on the upper side between 09/11/2011 and 17/05/2012 shows great improvement in the mortality rate. An interception on the upper side means that a lesser number of people died from the stroke in the hospital than expected results. Improvement requires investigation to find out the reasons for the positive growth and to maintain the same. The second interception is after16/07/2014 on the lower side showing that mortality due to stroke is worsening. The rise in mortality levels needs an investigation to find out the cause and make necessary changes. Issues evident when analysing the graph The graph clearly shows different phases in the improvement of the mortality levels due to stroke in the hospital. Mortality is an important indicator in showing healthcare quality (Doran, 2011). The rise in the slope in 2011 to 08/03/2013 past the point of interception shows improvement will less death due to stroke. The part that showed great improvement is before interception due to positive growth. This phase needs investigation so that the practices at the time may be doubled to get good results. The phase between 08/03/2013 to 2014 marked the period when the hospital experienced worsening mortality due to stroke. The period needs an investigation to find out practice changes that took place to cause the situation to become worse. The period has a downward slope in the graph associated with high mortality. The interception shows the out-of-control state in 2014 due to high mortality. The periods when the VLAD line does not intercept the limit lines shows that the mortality rate in the hospital has a similarity to national mortality rate. The improvements then fall in mortality levels at specific period need the hospital management to analyse the factors causing the effects. The factors may arise from many things including care to the patients. The way management treats workers and staffing also may play a great part in improving the service to the patients. Other causes may be the drugs used to treat the patients and patients nutrition changed after 08/03/2013. Painkillers form one of the main causes of death among patients suffering from a stroke. Hence, the hospital management should look into all factors that change during this time and reverse to the previous practices. Investigation and comparing all these factors will help in improving the quality of healthcare to patients and limit the number of mortality. Underlying issues The graph shows definite changes between two phases in mortality levels that occurred in different regions. It is clear that there is gradual worsening of the mortality levels in the hospital. As the head of the area, there is need to find out the cause of more deaths than in the previous time due to stroke. The cause of mortality getting worse may be due to many factors. Some of the factors include wrong entries of data, change in dealing with staff and levels of staffing and change in the type of medication for stroke patients. Another area of concern is the care services to the patients especially in the acute phase (Sabín, 2008) and availability of equipment to handle the patients. The investigation should use the pyramid model of investigation. The method uses the hierarchical approach to identifying the cause within five areas. The investigation starts from data at the bottom that has higher chances of the cause and finishes with the professionals. Actions to investigate the issues Data Under this level, the investigation determines if the data coding is accurate. Checking the data gives the first clue to the possible causes for the rise in mortality rate than the national average level in the hospital. The outcome improves the data quality to improve the credibility of the indicators. Case mix This level checks for variations that are usually not accounted for in the report. The level provides all risk adjustment factors that were not present at the data level (Duckett, Coory & Sketcher-Baker, 2007). It reveals some of the reasons for increased mortality rate that may be about patients’ conditions. This level shows if certain health conditions lead to an increase in the negative outcome. Structure and resources The level investigates the availability of required structures to handle the patients. The structures include hospital wards and the equipment to handle patients effectively. The level also investigates the staffing of the professionals and if some may become less efficient after working for long hours. Process of care The level includes an investigation on changes in services to the patient regarding the time when there was good mortality rate. There is also an investigation of changes in the frequency of providing such care to the patient. Besides, there is an investigation on variation in the process of providing the care at this level to show contributions to the negative outcome. Professional The investigation at this level looks into the changes in the staffs the wards dealing with stroke patients. Factors leading to negative outcome may be due to lack of proper experience in the present personnel or the staff members becoming tired after working for long durations. Other factors may include motivation levels of the staff in providing the services and cause of such changes. The findings from the investigation will determine actions necessary in correcting any problem. Such action may include educating the personnel on how to handle patient and provide sessions for interaction to provide more experience. There is a need to provide necessary resources and make changes in dealing with health care providers to increase the safety of the patients. Further, follow-up using VLAD since it is applicable in the short-term outcome (Pagel et al., 2012). Strategies to monitor Monitoring strategies involve regular audit after implementing the necessary changes. I will have audits done in a more frequent manner to note any change in mortality levels. A frequent audit may be through automation of the reporting system (Vasilakis, Wilson & Haddad, 2011). Furthermore, there will be noting any changes in the services to the patients and evaluate the effect after the shortest period. I will also check on the updates in guidelines to know any new requirements. Finally, I will keep up to date all health workers about any changes and getting their views on the alterations. Response to quality manager on patients’ experience Getting feedback from the patients forms one of the best ways of knowing that the hospital provides quality services. The recommendation from the quality manager is good in ensuring that the patients receive satisfactory services from the workers. The feedbacks may point specific points in correcting how to handle the patient. Besides, these feedbacks are vital to realising and making fast changes to improve the quality of the services since most of the feedbacks are instantaneous. Hence, the consideration into looking at the patients is important in improving service quality (McSherry, McSherry & Watson, 2012). References Andrianopoulos, N., Jolley, D., Evans, S. M., Brand, C. A., & Cameron, P. A. (2012). Application of Variable Life Adjusted Displays (VLAD) on Victorian Admitted Episodes Dataset (VAED). BMC Health Services Research, 12(3), 1-7. doi:10.1186/1472-6963-12-278 Browse, L. Black, J. Burnand, K. Corbett, S. & Thomas, W. (2010). Browses Introduction to the Investigation and Management of Surgical Disease. Boca Raton: CRC press. Doran, D. (2011). Nursing outcomes: The state of the science. Sudbury, MA: Jones & Bartlett Learning. Duckett, S. Coory, M. & Sketcher-Baker, K. (2007). Identifying variations in quality of care in Queensland hospitals. Med J Aust, 187 (10): 571-575 Faltin, F. W., Kenett, R., & Ruggeri, F. (2012). Statistical methods in healthcare. Chichester, West Sussex, United Kingdom: Wiley Foltran, F., Baldi, I., Bertolini, G., Merletti, F., & Gregori, D. (2009). Monitoring the performance of intensive care units using the variable life-adjusted display: a simulation study to explore its applicability and efficiency. Journal Of Evaluation In Clinical Practice, 15(3), 506-513. doi:10.1111/j.1365-2753.2008.01052.x Healy, J. (2013). Improving Health Care Safety and Quality: Reluctant Regulators. Farnham: Ashgate Publishing, Ltd ITCH (Conference), McDaniel, J. G., & IOS Press. (2009). Advances in Information Technology and Communication in Health. Amsterdam: IOS Press Lawless, J. (2014). Statistics in Action: A Canadian Outlook. Boca Raton: CRC press. McSherry, W., McSherry, R., & Watson, R. (2012). Care in nursing: Principles, values and skills. Oxford: Oxford University Press. Pagel, C., Prost, A., Nair, N., Tripathy, P., Costello, A., & Utley, M. (2012). Monitoring mortality trends in low-resource settings. Bulletin Of The World Health Organization, 90(6), 474-476. doi:10.2471/BLT.11.092981 Pagel, Christina., Prost, Audrey., Nair, Nirmala., Tripathy, Prasanta., Costello, Anthony & Utley, Martin. (February 6, 2012). Monitoring mortality trends in low-resource settings. Web. June 9, 2015. Retrieved from http://www.who.int/bulletin/volumes/90/6/11-092981/en/ Queensland Health. (2014). Variable Life Adjustment Display (VLAD). Web. June 9, 2015. Retrieved from http://www.health.qld.gov.au/psu/vlad/ Roberts, G. Tang, C. Harvey M. & Kadirkamanathan, S. (2012). Real-time outcome monitoring following oesophagectomy using cumulative sum techniques. World J Gastrointest Surg, 4(10), 234-237. Sabin J. (2008). In-Hospital Mortality in Stroke Patients. Rev Esp Cardiol. 61:1007-9. - Vol. 61 Sorensen, R., & Iedema, R. (2011). Managing Clinical Processes. London: Elsevier Health Sciences APAC. Vasilakis, C. Wilson, A. & Haddad S. (2011). Automating the monitoring of surgical site infections using variable life-adjusted display charts. J Hosp Infect, 79(2), 119-24. Read More