Dear Editor

Milne and Hill suggest the 'Getting Research into Practice' programme may have contributed to the decline in surgery for glue ear that we have reported. We cannot, however provide any evidence to support this suggestion. The rate of decline in surgery throughout England was similar to that observed in 13 former health districts in the Oxford/Anglia region. The ratio of change in rates after 1992 over the change in rates beforehand was 0.88 (95%CI 0.86 to 0.90) in the 13 districts and 0.88 (95%CI 0.84 to 0.92) in England. When we consider Berkshire alone (West and East Berkshire districts combined) where GRiP was initially implemented, the ratio was 0.87 (95%CI 0.84 to 0.90).

Dear Editor

Watkins and colleagues[1] add to the growing literature on the many and varied influences on general practitioner prescribing behaviour and costs.[2,3] They argue, from the results of their analyses, that the dispensing status of the general practitioner is a statistically significant predictor associated with high prescribing costs, though it is unclear from their report exactly how many dispensing practices are involved and the numbers in each prescribing cost category. One difficulty with the attribute of dispensing status is that practices are classed as such if they dispense for just a small proportion of their registered population or if they dispense to all the registered population, and we do not know to what extent this variable influences prescribing costs.

The authors point out that dispensing doctors, because of the need to operate their dispensary as a business, are likely to have better knowledge of drug costs than their non-dispensing colleagues. In addition, they are more likely to restrict repeat prescribing quantities to 28 days’ worth, as opposed to the two or three months’ worth that non-dispensing practices commonly issued at the time of the study. In theory this greater overall control over repeat prescribing should exert downward pressure on costs by controlling medicine wastage arising from patient demand for unnecessary repeat medication. Work carried out in Cornwall and Isles of Scilly supports the notion that the prescribing costs of dispensing practices are similar to, if not slightly less than, those of non- dispensing practices.[4] In addition, aggregate data from the Prescription Pricing Authority show, for Cornish practices, that the cost per person of prescriptions arising from dispensing practices is consistently lower than that of prescriptions originating from non- dispensing practices, and this has been the case for many years.

References

(1) Watkins C, Harvey I, Carthy P, Moore L, Robinson E, Brawn R. Attitudes and behaviour of general practitioners and their prescribing costs: a national cross sectional study. Qual Saf Health Care 2003;12:29-34.

(2) Jacoby A, Smith M, Eccles M. A qualitative study to explore influences on general practitioners’ decisions to prescribe new drugs. Br J Gen Prac 2003;53:120-125.

(3) Jaye C, Tilyard M. A qualitative comparative investigation of variation on general practitioners’ prescribing patterns. Br J Gen Prac 2002;52:381-386.

(4) Wilcock M. Dispensing doctors and non-dispensing doctors - a comparison of their prescribing costs. Int J Pharm Prac 2001;9:197-202.

Dear Editor

Regarding the article by L Locock,[1] we would like to share our healthcare redesign experiences with you. In recognising that up to 19.4% of all adverse patient events are attributable to medication mishap, Bayside Health Service (Melbourne, Australia) has identified medication safety as a priority.[2-5] A multidisciplinary Drug Management Committee was established to identify systemic factors contributing to medication error and implement safety strategies. Our experiences have demonstrated that the application of total quality management (TQM)/ continuous quality improvement (CQI) and re-engineering principles can effect significant medication process and practice improvement.

Anticoagulants contribute up to 11.2% of all medication-related adverse events.[2,3] Consultation with medical staff and patients identified that the requirement to prescribe warfarin for 2000 hours administration contributes to dosing complications, since the prescription is often written after-hours by staff who may be unfamiliar with the patient. To address this issue, warfarin administration time was altered to 1600 hours. This ensures that the treating practitioner doses his own patient rather than deferring this responsibility to after-hours staff; nursing staff are also prompted to contact the treating practitioner if warfarin has not been prescribed by the time it is due for administration (1600 hours). The change was implemented hospital-wide over a brief period, and embraced by both clinical staff and patients. This sustainable change has resulted in reduced potential for bleeding complications and improved workflows for medical, nursing and pathology staff.

Infusion rate errors are also commonly reported in the literature, and contribute to approximately 30% of all medication incidents reported at Bayside Health.[6] A review of intravenous administration guidelines for 13 high-risk medications (e.g. dopamine) identified considerable variation that was likely to cause confusion and error. Infusion guidelines were standardised for 12 of 13 high-risk medications (92%). In contrast to the warfarin initiative, this change process necessitated lengthy consultation and negotiation over a year.

Both TQM/CQI and re-engineering approaches have been subject to criticism regarding impact and sustainability.[7] However, Bayside’s two redesign initiatives demonstrate the effectiveness of strong executive support and leadership, the need to create a sense of urgency and a common vision, and to demonstrate early successes to engage staff.[8] The importance of active clinician participation from the outset is critical to the success of any change strategy; staff must believe that the proposed practice change will contribute to improved workflows and patient safety. Whilst it may be argued that healthcare quality improvement necessitates major structural reform, our experiences indicate that simple systems redesign is effective in achieving positive change.[9]

References

(1) L Locock Healthcare redesign: meaning, origins and application. Qual Saf Health Care 2003;12: 53-57.

(2) Leape LL, Brennan TA, Laird NM et al. The Nature of Adverse Events in Hospitalized Patients: Results of the Harvard Medical Practice Study II. NEJM 1991; 324 (6): 377-384.

(3) Wilson RM, Runciman WB, Gibberd RW et al. The Quality in Australian Health Care Study. MJA 1995; 163: 458-471.

(4) Kohn LT, Corrigan JM, Donaldson MS. To err is human: building a safer health system. Committee on Quality of Health Care in America. Institute of Medicine. Washington, DC: National Academy Press, 1999.

(5) Australian Council for Safety and Quality in Health Care, National Action Plan 2002. Available at: http://www.safetyandquality.org/articles/Publications/actplan02.pdf

(6) Flynn F, Mohr L, Lawlor-Klean P. Right Programming of Pumps to Prevent Errors in the Infusion Process. Joint Commission Journal on Quality and Safety 2003; 29(1) 37-40.

(7) Locock L. Healthcare redesign: meaning, origins and application. Qual Saf Health Care 2003;12: 47-52.

(8) Kotter JP. Leading Change: Why Transformation Efforts Fail. Harvard Business Rev 1995, 59-67.

(9) Nolan TW. Understanding Medical Systems. Ann Intern Med 1999; 128:293-298.

Dear Editor

We need to do a better job of keeping track of potential side effects when designing randomized clinical trials (RCTs). Consider a RCT for a new drug tested for the main, or first order effect, the reduction of hypertension. Power calculations are carried out so that meaningful differences between the drug users and the controls can be detected on this effect. Second order effects, mortality in this case, are those adverse events that are prospectively tracked, recorded, planned for in advance, and if sufficiently serious may bring the trial to a halt through review conducted by a Data Safety Monitoring Board. Too rarely are there systematic prospective efforts to collect information on other unpredicted effects of the drug being tested. These third order effects may be important, but are rare and unexpected and thus not necessarily sought out and consequently systematic data are not necessarily prospectively collected about them. This problem of falling to track such third order events is vividly shown in the recent article about Bruce Psaty and his recognition of the 3rd order side effects related to the use of calcium channel blockers [1].Although the United State of America Food and Drug Administration (FDA) and other Government agencies have systems in place for the post-marketing reporting of possible side effects, this surveillance is passive in nature and relies mainly on the judgment of practitioners to suggest whether or not a patient's complaint is related to the drug in question. This creates an inherited underreporting of side effects. It also assumes that everyone has the same access to health care or that by the time the complaint is reported there is enough time to avoid serious complications in other patients.

The USA National Institutes of Health (NIH) and the World Health Organization (WHO) are aware of these issues. NIH guidelines appear to be focused on multi-center clinical trials. The definition of adverse events less well characterized for single center trials. Furthermore, the definition of an adverse event, while in compliance with the FDA regulations, is somewhat broad or vague. For example, the federal regulation requires the reporting of adverse events that are "serious and unexpected". This term could be interpreted differently. Does the event in question need to be both serious and unanticipated by the investigator? What about unanticipated effects regardless of the seriousness? How serious? Are events that are costly to a patient in terms of productive time, money or psychological damage considered serious? See Table 1 for a summary of the NIH Institutes and their guidelines and their web sites.

This lack of uniformity and clarity is not particular to the United States. Some international institutions rely on the World Health Organization (WHO) to establish their adverse reporting mechanism standards. The WHO in their Guidelines for good clinical practice (GCP) for trials on pharmaceutical products, states that the investigator is "…responsible for notifying (with documentation) the relevant health authorities, the sponsor and, when applicable, the ethics committee immediately in the case of serious adverse events or reactions, as governed by national regulations…" This statement contains little guidance as to who is the relevant authority, what documentation to use and under what circumstances the established ethical committee should be involved. Furthermore, the statement "…as governed by national regulations", does not clearly assign responsibility for the reporting of adverse events, especially in situations when one nation is conducting medical trials in another country. Moreover the WHO falls short in suggesting any type of analyses for the examination of possible population wide repercussions[2].

Countries in Central and South America basically follow the guidance of the WHO through the Pan-American Health Organization (PAHO). In England and Australia, countries that have a comprehensive system of health care, information about adverse events, especially if observed after the trials and/or the medication is in the open market, is not easily accessible.

We think there should be more clarity and uniformity in these guidelines and all guidelines should be accessible through the Internet or some other form of common global communication to the scientific community. Furthermore, investigators and agencies alike should be aware of studies reporting adverse events. These reporting mechanisms should have direct relationship with post marketing surveillance efforts [3, 4].

Table 1 Procedures for Monitoring Adverse Events for USA NIH Institutes

1: NIH: National Institutes of Health; NCI: National Cancer Institute; NEI: National Eye Institute; NHLBI: National Heart, Lung and Blood Institute; NHGRI: National Human Genome Research Institute; NIA: National Institute on Aging; NIAAA: National Institute on Alcohol Abuse and Alcoholism; NIAID: National Institute of Allergy and Infectious Diseases; NIAMS: National Institute of Arthritis and Musculoskeletal Diseases; NIBIB: National Institute of Biomedical Imaging and Bioengineering; NICHD: National Institute of Child Health and Human Development; NIDCD: National Institute on Deafness and Other Communication Disorders; NIDCR: National Institute of Dental and Craniofacial Research; NIDDK: National Institute of Diabetes and Digestive and Kidney Diseases; NIDA: National Institute on Drug Abuse; NIEHS: National Institute of Environmental Health Sciences; NIGMS: National Institute of General Medical Sciences; NIMH: National Institute of Mental Health; NINDS: National Institute of Neurological Diseases and Stroke; NINR: National Institute of Nursing Research.

References (1) Deyo RA. Bruce Psaty and the risks of calcium channel blockers. Qual Saf Health Care 2002 11(3): 294-6.

(2) WHO. Guidelines for good clinical practice (GCP) for trials on pharmaceutical products. 1995. p. Annex 3.

(3) Begg C et al. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. .JAMA 1996. 276(8):637-9.

(4) Tramer MR et al. Quantitative estimation of rare adverse events which follow a biological progression: a new model applied to chronic NSAID use. Pain 2000. 85(1-2):169-82.