Quantitative Research in Patient Safety – Literature Review

Recently, research has occupied a crucial place in nursing that is identified as “the diagnosis and treatment of human responses to actual or potential health problems” (American Nurses’ Association, 1980 p.9); thus, an appropriate understanding of research literature is a prerequisite for every individual who works in this area (Rees, 2003). Unfortunately, despite the fact that most of nurses acquire specific skills in research, only some of them manage to apply research data or research findings to practice (Bostrum & Suter, 1993). This can be explained by the nurses’ inability to critique a research, evaluating its pros and cons (Krainovich-Miller et al., 2002). The aim of the present essay is to critically analyse two quantitative research literatures in patient safety. The first research is “Relationship between complaints and quality of care in New Zealand: a descriptive analysis of complainants and noncomplainants following adverse events” by M. Bismark et al. (2006), while the second research is “Comparison of three methods for estimating rates of adverse events and rates of preventable adverse events in acute care hospitals” by P. Michel et al. (2004). Although many nursing studies have been conducted in the last decade (e.g. Johnson & Lauver, 1989; Conlon & Anderson, 1990; Norman et al., 1991; Brennan et al., 1995; Gross et al., 1995; Fieler et al., 1996; Bennet, 1999), they implicitly dealt with the issues of patient care; however, the studies discussed further directly relate to the quality of medical care in New Zealand and France.

The research conducted by Bismark et al. (2006) evaluates the extent of injuries in the patients cured in public hospitals of New Zealand, or more precisely (as the title reveals), a correlation between patients’ complaints and quality of medical care. While the title is clue to the focus, the abstract gives more detailed information, identifying the major aspects of the research (objectives, design, setting, population, main outcome measures, results and conclusion) in a clear scientific style. However, the abstract does not indicate the research questions of the study; they are stated further in the research and are the following:

1) Do complaints track injuries, or are they prompted by more subjective concerns?

2) Are complaints the “tip of the iceberg’ in terms of quality of care problems and, if so, how representative are they of broader quality problems? (Bismark et al., 2006 p.17).

Although the research by Michel et al. (2004) also refers to patient safety, neither the title, nor the abstract uncovers the theme in an explicit way. Actually, the theme is exposed further in the research; in particular, the study analyses rates of unintended injuries (defined by the authors as adverse events and preventable adverse events) in the patients cured in care hospitals of France. Similarly to the first research, the abstract in the second study briefly summarises the research and is divided into the same categories that uncover the essence of the investigation. In this regard, the abstract is an obvious strength of the analysis and it can serve as an example to other researchers who investigate various aspects of nursing.

But the research does not specify the research questions either in the abstract or in the introduction section of the paper. Such a lack of specific questions certainly complicates the overall apprehension of the study. The authors could have proposed some research questions, such as

1) What are the major aspects of reliability, acceptability and effectiveness?

Or

2) How rates of adverse events and rates of preventable adverse events can be properly assessed with each of three methods?

These questions are of primary importance to the research, as adverse events and preventable adverse events can not be rightfully evaluated, if the major criteria of reliability and effectiveness are not properly discussed in the context of the research. However, the authors pay little attention to these aspects of the analysis.

Despite the fact that the introduction section in both studies provides a valid explanation of the importance of the problem, neither of the two studies includes an overview of the previous research or specific reports. This neglect decreases the overall presentation and reduces the value of the presented data. However, the problems of statement are formulated in a concise way and reflect that the researchers narrowed the areas of research to the issue of adverse events in the clinical setting in order to get more accurate findings. In fact, this issue is especially relevant today when patient safety has become worse in many countries of the world. The justification for the chosen topic in the research by Bismark et al. (2006) is that the recent accident compensation system in New Zealand does not adequately examine patients’ complaints in all cases of adverse events. Pointing at the fact that “there is growing international interest in harnessing patient dissatisfaction and complaints to address problems with quality” (Bismark et al., 2006 p.17), the authors concurrently put crucial questions that inspire readers’ interest in the issue of patient safety from the very beginning. In the research of Michel et al. (2004) the underlying reason for initiating an investigation is that the limitations of the employed methods reduce the validity of the received findings in regard to patients’ injures within the hospital setting. However, the lack of appropriate background, theoretical frameworks, hypotheses and definite aims in the introduction section considerably limits the studies. This especially regards the non-inclusion of specific theories that usually back up the presented data. In this respect, both studies are theory-free; unlike theory-testing research and theory-generating research, this kind of research is less popular because it does not analyse any theoretical concepts that constitute the basis of practical nursing. On the other hand, the studies of Michel et al. (2004) and Bismark et al. (2006) specifically focus on a practical problem-solving framework; that is, the present researches are aimed at identifying practical solutions to the discussed problems rather than discussing theoretical implications.

The research of Michel et al. (2004) uses a quantitative research method that “emphasizes objectivity through statistical analysis” (Santy & Kneale, 1998 p.77) and the quasi-experimental design that is considered to be more adequate and less biased than an experimental method, if an investigation is conducted within the clinical setting (Polit & Hungler, 1995). Though objectivity is crucial for such kind of research, it would also be appropriate to combine quantitative and qualitative methods, that is, to combine objectivity and subjectivity (Phillips, 1990). The fact is that due to its quantitative method the study appears to be too analytical, too objectively-oriented; thus, there is a necessity to introduce some aspects of the subjective realm into the research. However, Parahoo (1997) supports another viewpoint, exposing the inadequacy of a qualitative method, especially in regard to a nursing research. The author points out that, applying to a quantitative method, researchers are able to predict the final outcomes, while a qualitative method may generate unpredictable results. The data in the study are collected in care hospitals of Aquitaine with the help of three research techniques – a cross sectional method, a prospective method and a retrospective method. Such triangulation is aimed at “relat[ing] different sorts of data in such a way as to counteract various possible threats to the validity of analysis” (Hammersely & Atkinson, 1983 p.199). In the present study triangulation corresponds with the terms of reference that provide appropriate relevance to the whole research (Shih, 1998). Identifying both advantages and disadvantages of all three methods in Box 2, the researchers contribute much to the reliability of the findings, despite the fact that they have not conducted a pilot study that, according to Carr (2003), intensifies the credibility of the employed research techniques. On the other hand, a pilot study is crucial for the investigations that utilise unchecked tools for research, as is the case with the present study, where the researchers conduct an evaluation of methodology. In this regard, a pilot study “helps to illuminate some of the problems of the research tool” (Santy and Kneale, 1998 p.80).

The research of Bismark et al. (2006) is also quantitative with descriptive design. The baseline data are taken from the medical records of the New Zealand Quality of Healthcare Study (NZQHS) and the Commissioner’s complaints database. Further, multivariate and bivariate analyses are applied to the research to identify certain dissimilarities between complaints and non-complaints. Overall, the explanation of the research techniques and methods is a great strength of this study, as the authors provide a thorough description in regard to data collection and study design. Although the researchers do not define a hypothesis of the analysis, they, nevertheless, use dependent and independent variables to differentiate complainants from non-complainants. However, the limited space of both studies has not allowed the researchers to insert the samples of medical records and questionnaires that served as the basis for the research; thus, the methodology of both investigations can not be fully assessed in terms of the quality.

Actually, the research of Bismark et al. (2006) and the research of Michel et al. (2004) employ primary sources (including official records) that explicitly relate to the subjects. But according to Burgess (1991), even primary sources should be critically assessed and “it is essential to locate them in context” (p.124). But neither the first nor the second study provides a critical evaluation of the utilised sources.

In regard to ethical issues, they are not openly addressed in the studies; however, in the research of Bismark et al. (2006) there is a mentioning that the investigation was endorsed by the Wellington Ethics Committee. For Robinson (1996), such ethical approval is a necessary part of a nursing research, as any investigation deals with human beings who may experience certain difficulties during the research. On the other hand, due to its descriptive nature the present study does not necessarily need an informed consent or ethical considerations (Cutcliffe & Ward, 2003), while the research of Michel et al. (2004) requires a discussion of certain ethical issues because of its quasi-experimental design. Some of these issues are patients’ confidentiality, defence of their rights and risk control (Pranulis, 1996). In regard to the latter factor, it is necessary for researchers to increase potential benefits and decrease potential risks, especially in such studies that involve a great number of participants, as is just the case with the research of Michel et al. (2004). Thus, it would have been proper for the researchers of the present study to discuss in detail subjects’ conditions and potential harm, particularly in view of the fact that nursing directly relates to patient safety within the clinical setting (DHHS, 1981). However, the ethical rights of samples are implicitly defended in both studies, as no personal details of participants are revealed. But the researchers provide no information of the ways the data were stored and protected before or during the investigation. Similarly, neither of the studies refers to informed consents, while this is a prerequisite for any nursing research (Alt-White, 1995; Berry et al., 1996).

As for sampling, the study of Bismark et al. (2006) analyses two groups of patients: the first group includes people who made complaints to the Commissioner and the second group includes people “identified by the NZQHS as having suffered an adverse event who did not lodge a complaint” (Bismark et al., 2006 p.17). A two stage sampling process is initiated by NZQHS on the example of 6579 medical records. Although inclusion and exclusion criteria are not explicitly identified in the study, the researchers make it clear that they only choose the patients who suffer adverse events. In the process of analysis these patients are divided into two categories – complainants and non-complainants, though both groups are typical representatives of the larger population. In the research of Michel et al. (2004) the sampling includes 778 patients from medical, surgical and obstetric wards. This number of samples is appropriate for a descriptive study.

Initially, the researchers chose 786 patients with the help of a two stage cluster stratified process, but excluded 8 persons “because they were still present on day 30, precluding the review of their medical records” (Michel et al., 2004 p.2). In this respect, the study does not clearly define inclusion and exclusion criteria, but some samples are excluded in the process of investigation. No obvious bias is found in regard to the samples; similar to the previous research, the samples belong to typical representatives of the larger group. In view of this fact, the sampling can be considered as fully reliable.

In addition to authors’ comments, the results in the research of Bismark et al. (2006) are presented in figures, tables and boxes that are introduced as additional tools for clarification. This visual information reflects how the data are collected and measured (Figure 1 is especially accurate in revealing the cases of injured complainants and non-complainants). Although the authors do not specifically explain such a choice, they provide a detailed justification for the use of correlation tests that define dependent variables (a distinction between complainants and non-complainants) and independent variables (age, ethnicity, sex and other factors). Besides, the researchers weight the bivariate and multivariate analyses to acquire more accurate findings. The results in the study of Michel et al. (2004) also appear in both textual and graphic forms in order to enhance explanation. But the researchers do not attain the balance between figures and comments, putting too much emphasis on figures. Unlike the previous study, the authors do not use dependent and independent variables in their analysis; however, they employ paired X² tests for the comparison of retrospective and prospective methods.

Discussing their findings, Bismark et al. (2006) draw a parallel between the received results and the findings of the previous studies. Actually, many findings of the prior research are consistent with the present research (e.g. Burstin, et al., 1993; Studdert et al., 2000), while some findings contradict the earlier results (e.g. Tapper et al., 2004). To some extent, such a comparison justifies the lack of literature review at the beginning of the research and provides more validity to the overall outcomes. In general terms, the findings of Bismark et al. (2006) directly relate to the objectives of the study, gradually introducing the evidence that proves the authors’ initial suggestions. In particular, the researchers find out that 79% of all injures can be identified as preventable adverse events. In the case of the Commissioner’s analysis, 64% of the complaints are made by the patients who suffer adverse events, of which 51% are preventable adverse events. In regard to the NZQHS review, 315 cases of adverse events (out of 850 cases) are preventable, 124 cases are serious and 48 cases are serious and preventable. As for instigators of complaints, 41% of complaints are made by the patients, while 59% – by their relatives or friends (13% – spouse, 16% – parent and 17% – child).

Evaluating the independent variables, the researchers reveal that the age of complainants is lower than the age of non-complainants; moreover, non-complainants mainly live in the regions with poor economic conditions. The findings in the research of Michel et al. (2004) also relate to the terms of reference, providing evidence that “the prospective method has several advantages over retrospective and cross sectional methods” (Michel et al., 2004 p.3). In particular, the prospective method better recognises preventable adverse events and is more trustworthy than two other methods. This is clearly seen in Venn diagrams that demonstrate the number of adverse events identified by each of three research methods. Overall, the findings in the present study are not properly discussed; however, the researchers discuss in detail the strengths and limitations of the research in the discussion section. For instance, as the authors reveal, reliability and effectiveness of adverse events rates are successfully estimated because the samples are assessed with the help of three methods. On the other hand, the researchers point at the possibility of bias that “may have been present due to the small number of hospitals and wards” (Michel et al., 2004 p.3) and because of the participation of the care teams in the prospective method. Besides, the reference list that the researchers utilised in the process of investigation might have errors that were not identified. Finally, the aspects of reliability, effectiveness and acceptability are not discussed in detail by the authors, though these are the major assessment criteria of the study. However, there are some obvious strengths of the research; unlike the studies that analyse adverse events either in surgery or medicine (e.g. Mantel et al., 1998; Waterstone et al., 2001), the present study examines various cases of adverse events in three areas – medicine, surgery and obstetrics.

Similarly to this research, Bismark et al. (2006) also identify certain limitations of the study; in particular, the analysis of adverse event rates is rather confined, if medical record reviews serve as the basis for the research. The research also lacks definite ethnicity data for all complainants; thus, there is a “potential for measurement error” (Bismark et al., 2006 p.21). In addition, the authors do not provide any information as to the alternative research methods that can be used for the assessment of the relations between complains and quality of medical care. Drawing a parallel between the employed methods and the alternative methods, it will be possible to enhance the validity of the received findings. The conclusion in the research of Bismark et al. (2006) directly responds to the terms of reference; based on the received results, the conclusion suggests that elderly or economically poor patients rarely initiate complaints processes. The same regards the patients who belong to ethnic minorities (in this case – to Pacific ethnicity). The authors recommend to conduct a further study that will profoundly investigate the reasons for people’s refusal to make complaints in the cases of poor medical care. Moreover, the complaints greatly depend on the severity of injures and whether the event is preventable or unpreventable.

In this respect, as the researchers conclude, “complaints offer a valuable portal for observing serious threats to patient safety and may facilitate efforts to improve quality” (Bismark et al., 2006 p.22). Unfortunately, no recommendations for practice are made at the end of the study, thus reducing the relevance of the received findings. On the other hand, as Santy and Kneale (1998) claim, “all research has some implications for practice even if the results have proven to be inconclusive” (p.82). In the research of Michel et al. (2004) the conclusion summarises the results that, in the authors’ words, “provide new insights into the epidemiology of adverse events” (p.4). Such a viewpoint is explained by the fact that the findings of the present study reveal the ways to intensify the implementation of prospective assessment in the clinical setting. However, the researchers only suggest the answers to the posed questions, avoiding any insistence on specific concepts or notions. Comparing three methods, the researchers recommend to use the prospective method for different purposes that implicitly or explicitly relate to the evaluation of adverse events rates. Finally, Michel et al. (2004) briefly discuss the prior knowledge on the topic and the knowledge acquired in the process of investigation. In regard to the prior knowledge, the assessment of adverse events was conducted in an analytical way that considerably limited the findings. In the present study the researchers receive more feasible results and identify that the causes of adverse events and risk reduction programmes can be successfully evaluated by the prospective method rather than by the retrospective or cross-sectional methods. However, further research is required, if the evidence received in this research is applied to practice (Barron & Kenny, 1986; Scott & Thompson, 2003).

Overall, both researches are well-structured and are written in a scientifically concise style; however, as was stated above, the study of Michel et al. (2004) provides too much technical details, while analysing the results. Therefore, it is slightly difficult to read the research and, consequently, there is a chance that its findings may be ignored by a practitioner on the premise of misunderstanding. Although the research of Michel et al. (2004) is logically constructed, an unqualified person may fail to rightfully apprehend the presented data. On the contrary, the study of Bismark et al. (2006) is easy to understand because it lacks much unexplained jargon. Another strength of the research is the appropriate use of quotes in the discussion section; these quotes are directly related to the analysis and correspond with the ideas expressed by the authors, either refuting or confirming them. Employing this or that quote, the researchers provide a detailed interpretation of a certain concept; and for all that, the number of quotes is reasonable and they are rather short. On the contrary, Michel et al. (2004) do not utilise quotes in the discussion to support their arguments, though they use certain references. Despite the fact that the researchers do not explicitly recommend their studies to nurses, the overall findings can be especially relevant to nursing staff, as well as to the researchers who are involved in health care.

Within a complex clinical setting nurses experience various difficulties because of the lack of appropriate practical knowledge (Treacy & Hide, 1999; Polit et al., 2001). Thus, the studies of Michel et al. (2004) and Bismark et al. (2006) can inspire nurses’ interest in the ways of patient safety, as, despite their limitations and certain inadequacies, the studies pose vital questions that may increase the quality of medical care not only in France and New Zealand, but in other countries as well. Due to the fact that nowadays nursing staff is usually required to implement various aspects of research into practice (Christman & Johnson, 1981; Burnard & Morrison, 1990; Street, 1992; McSherry, 1997; Cormack, 2000; Rodgers, 2000; Hek et al., 2002; Cluett & Bluff, 2004), the present studies are especially valuable, as they provide useful and valid information that extends the prior knowledge in patient safety. In further studies it will be crucial to discuss the received findings in the context of international implications and to pay more attention to preventable adverse events (Thomas et al., 2000). Moreover, it will be important to give some recommendations for nursing staff and those individuals who deal with patients’ complains (World Health Organisation, 1977; Gordon, 1988; Brink et al., 1989; Lindley & Walker, 1993; Ferketich, & Mercer, 1995; Northouse, 1995; Roseman & Booker, 1995; Duffy et al., 1996; Madge et al., 1997; Vertanen, 2001).

Bibliography

Alt-White, A. C. (1995) Obtaining ‘informed’ consent from the elderly. Western Journal of Nursing Research, 17, 700-705.

American Nurses Association (1980) Nursing: A Social Policy Statement. Kansas City, American Nurses Association.

Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Bennet, J. A. (1999) Critique of research review of attitudes to HIV/AIDS. Journal of Advanced Nursing, 29 (3), 770-771.

Berry, D. L., Dodd, M. J., Hinds, P. S., & Ferrell, B. R. (1996) Informed consent: Process and clinical issues. Oncology Nursing Forum, 23, 507-512.

Bismark, M. M., Brennan, T. A., Paterson, R. J., Davis, P.B., & Studdert, D. M. (2006) Relationship between complaints and quality of care in New Zealand: a descriptive analysis of complainants and noncomplainants following adverse events. Quality and Safety in Health Care, 15, 17-22.

Bostrum, J. & Suter, W. N. (1993) Research utilisation: making the link with practice. Journal of Nursing Staff Development, 28-34.

Brennan, P.F., Moore, S.M., & Smyth, K.A. (1995).The effects of a special computer network on caregivers of persons with Alzheimer’s disease.Nursing Research, 44, 166-172.

Brink, C.A., Sampselle, C.M., & Wells, T.J. (1989). A digital test for pelvic muscle strength in older women with urinary incontinence.Nursing Research, 38, 196-199.

Burgess, R. G. (1991) In the Field: An Introduction to Field Research. London, Routledge.

Burnard, P. & Morrison, P. (1990) Nursing Research in Action: Developing Basic Skills. London, Macmillan.

Burstin, H. R., Johnson, W.G., Lipsitz, S. R. et al. (1993) Do the poor sue more? A case control study of malpractice claims and socioeconomic status. JAMA, 270, 1697-1701.

Carr, J. (2003) Improving questionnaire response rates. Practice Nursing, 14 (4), 171-174.

Christman, N. J. & Johnson, J. E. (1981) The importance of research in nursing. In: Y. M. Williamson (ed.) Research Methodology and Its Application in Nursing. New York, Wiley. pp.3-24.

Cluett, E. R. & Bluff, R. (2004) Principles and Practice of Research in Midwifery. London, Bailliere Tindall.

Conlon, M., & Anderson, G. (1990). Three methods of random assignment: Comparison of balance achieved on potentially confounding variables. Nursing Research, 39, 376-379.

Cormack, D. (2000). The Research Process in Nursing. Oxford, Blackwell Science.

Cutcliffe, J. R. & Ward, M. (2003) Critiquing Nursing Research. Bath, Bath Press.

Department of Health and Human Services (DHHS) (January 26, 1981) Final regulations amending basic HHS policy for the protection of human research subjects. Federal Regulations, 46 (16).

Duffy, M. E., Rossow, R., & Hernandez, M. (1996).Correlates of health-promotion activities in employed Mexican American women.Nursing Research, 45, 18-24.

Ferketich, S. L., & Mercer, R. T. (1995).Paternal-infant attachment of experienced and inexperienced fathers during infancy. Nursing Research, 44, 31-37.

Fieler, V. K., Wlasowicz, G. S., Mitchell, M. L., Jones, L.S., & Johnson, J. E. (1996). Information preferences of patients undergoing radiation therapy. Oncology Nursing Forum, 23, 1603-1608.

Gordon, D. R (1988) Tenacious assumptions in Western biomedicine. In: Lock M, Gordon D. R (eds) Biomedicine Examined. London, Kluwer Academic Press. pp. 19–56.

Gross, D., Conrad, B., Fogg, L., Willis, L., & Garvey, C. (1995). A longitudinal study of maternal depression and preschool children’s mental health. Nursing Research, 44, 96-101.

Hammersley, M. & Atkinson, P. (1983) Ethnography: Principles in Practice. London, Tavistock.

Hek G., Judd, M., Moule, P. (2002) Making Sense of Research: An Introduction for Health and Social Care Practitioners. Sage Publications, London.

Johnson, J. E. & Lauver, D. R. (1989) Alternative explanations of coping with stressful experiences associated with physical illness. Advances in Nursing Science, 11 (2), 39-52.

Krainovich-Miller, B., LoBiondo-Wood, G. & Haber, J. (2002) Critical reading strategies: Overview of the research process. In: LoBiondo-Wood & J. Haber (eds.), Nursing Research: Critical Appraisal, and Utilization. St Louis, MO, Mosby. pp.33-50.

Lindley, P., & Walker, S. N. (1993).Theoretical and methodological differentiation of moderation and mediation.Nursing Research, 42, 276-279.

Madge P, McColl J, Paton J. (1997) Impact of a nurse-led home management training programme in children admitted to hospital with acute asthma: a randomised controlled study. Thorax, 52, 223–228.

Mantel, G. D., Biuchmann, E., Rees, H., Pattinson, R. C. (1998) Severe acute maternal morbidity: a pilot study of a definition for a near-miss. British Journal of Obstetrics and Gynaecology, 105, 985-990.

McSherry, R. (1997) What do registered nurses and midwives feel and know about research? Journal of Advanced Nursing, 25, 5, 985-998.

Michel, P., Quenon, J. L., Sarasqueta, A.M., Scemama, O. (2004). Comparison of three methods for estimating rates of adverse events and rates of preventable adverse events in acute care hospitals. British Medical Journal, 328, 1-5.

Norman, E., Gadaleta, D. & Griffin, C. C. (1991) A evaluation of three blood pressure methods in a stabilized acute trauma population. Nursing Research, 40, 86-89.

Northouse, L. L., Jeffs, M., Cracchiolo-Caraway, Lampman, L., & Dorris, G. (1995). Emotional distress reported by women and husbands prior to a breast biopsy.Nursing Research, 44, 196-201.

Parahoo, A. K. (1997) Nursing Research, Principles, Process, and Issues. London, MacMillan.

Phillips, D. C. (1990). Subjectivity and objectivity: An objective inquiry. In: Eisner and Peshkin (Eds.) Qualitative inquiry in education: The continuing debate (pp. 19-37). New York, Teachers College Press.

Polit, D. F. & Hungler, B. P. (1995) Nursing Research Principles and Methods. Philadelphia, J. B. Lippincott.

Polit, D. F., Beck, C. T. & Hungler, B. P. (2001) Essentials of Nursing Research Methods, Appraisal and Utilization. Philadelphia, Lippincott.

Pranulis, M. F. (1996) Protecting rights of human subjects. Western Journal of Nursing Research, 18, 474-478.

Rees, C. (2003) Introduction to Research for Midwives. London, Books for Midwives.

Robinson, J. (1996) It’s only a questionnaire: ethics in social science research. British Journal of Midwifery, 4, 41-46.

Rodgers S (2000) A study of the utilisation of research in practice and the influence of education. Nurse Education Today, 20 (4), 279-287.

Roseman, C., & Booker, J. M. (1995). Workload and environmental factors in hospital medication errors. Nursing Research, 44, 226-230.

Santy, J. & Kneale, J. (1998) Critiquing quantitative research. Journal of Orthopaedic Nursing, 2, 77-83.

Scott, T. J., & Thompson D. R. (2003) Assessing the information needs of post-myocardial infarction patients: a systematic review. Patient Education and Counselling. 50 (2), 167-177.

Shih, F. J. (1998) Triangulation in nursing research: issues of conceptual clarity and purpose. Journal of Advanced Nursing, 28 (3), 631-641.

Street, A. F. (1992) Inside Nursing: A Critical Ethnography of Clinical Nursing Practice. New York, State University Press of New York.

Studdert, D. M., Thomas, E. J., Burstin, H. R. et al. (2000) Negligent care and malpractice claiming behaviour in Utah and Colorado. Medical Care, 38, 250-260.

Tapper, R., Malcolm, L., & Frizelle, F. (2004) Surgeons’ experience of complaints to the Health and Disability Commissioner. New Zealand Medical Journal, 117 (1198), 1-122.

Thomas, E.J, Studdert, D. M, Burstin, H. R., Orav. E. J.,