Introducing patient safety to undergraduate medical students – A pilot program delivered by health care administrators

Abstract

Background: Identifying informed and interested staff to teach patient safety can be difficult. We report our experiences with a lecture-based program on patient safety delivered by health care administrators.

Method: A self-administered questionnaire survey on patient safety culture was given to 130 third-year medical students before (‘pre-test’) and at 3 months after (‘post-test’) the program. The latter consisted of two 60-minute whole-class lectures using contemporary medical incidents in this locality as illustrative cases.

Results: Thirteen of the 23 questionnaire items (56.5%) showed statistically significant changes at the post-test. Students became more appreciative of the notions that errors were inevitable, and that disciplinary actions and ‘being more careful’ may not be the most effective strategies for the prevention of error. Issues concerning error reporting were more resistant to our educational intervention. Students regarded patient safety as an important topic that should be included in undergraduate teaching and professional examinations.

Conclusion: The implementation of a patient safety curriculum should be tailored to an institution's needs, limitations, and culture. Health care administrators were found to be effective faculties. A brief lecture-based program could be integrated readily into an existing curriculum, and was found to produce at least short-term, positive changes.

Introduction

Medical incidents have become important educational resources in recent years and the introduction of patient safety into undergraduate medical education signifies a major change in culture (Halbach & Sullivan 2005; Patey et al. 2007; Anderson et al. 2009). Medical student teaching plays an important role in promoting the correct concepts and skills in patient safety but there is as yet no consensus on how the subject should be taught (Flanagan et al. 2004; Fischer et al. 2006). Moreover, advocates of patient safety in many health care systems are mostly administrative staff or non-academic clinicians who may not have the necessary skills or platform to teach students. Clinicians who have intuitively adopted patient safety methods in their practice may not know how to articulate what they do. Identifying informed and interested staff to teach patient safety may present as a significant teething problem.

These potential difficulties have been identified in the curriculum guide recently published by the World Health Organization (World Health Organization 2009). In addition to measures targeted at improving staff engagement, the WHO curriculum guide also suggests the involvement of other professionals during the implementation of a patient safety curriculum including engineers, behavioral scientists, and nurses. The potential role played by administrative personnel, however, has not been fully explored.

At our institution, we have introduced a program on patient safety for third-year medical students in collaboration with the Quality and Safety Division of the Hospital Authority (HA). The latter oversees public health care provision and is instrumental in promoting patient safety culture in this locality. We report herewith our experiences with a pilot program on patient safety delivered by health care administrators from the HA.

Table

Practice points

Implementing a program on patient safety can be challenging. An introductory program, however brief, is a good start. Health care administrators may be considered as effective faculties. A lecture-based program can be readily integrated into an existing curriculum. Some areas, such as the promotion of a healthy reporting culture, may require alternative methods of delivery.

Methods

The survey

A voluntary self-administered questionnaire survey was given to 130 third-year medical students. None of them had received any prior teaching on patient safety. The survey was given before the lectures in August 2009 (the ‘pre-test’), and after 3 months, in November 2009 (the ‘post-test’). Both were given during whole-class sessions in lecture rooms. Students had around 10 minutes to complete the questionnaire. The study was approved by the Institute of Health and Medical Sciences Education (IHMSE) of the university.

The questionnaire was adapted with permission from the investigators of a previous study (Madigosky et al. 2006). It consisted of 23 items – 15 addressed students’ perceptions on the nature and management of medical errors, three addressed their attitudes towards undergraduate teaching of patient safety, and five assessed their self-appraisals (not factual recall) of patient safety knowledge. Students’ responses were graded using a 5-point ordinal scale (1 = strongly disagree/very poor, 2 = disagree/poor, 3 = neutral/fair, 4 = agree/good, 5 = strongly agree/very good).

The paired differences of each student's responses at the pre- and post-tests were studied using the Wilcoxon signed ranks test. The overall direction and degree of change for each item was generated. Findings from 15 items concerning the nature and management of errors, and five knowledge items were reviewed by our faculties for ‘intended changes’. The latter were defined as what the faculties had intended to change with their teachings in accordance with the current thinking on patient safety. Statistical analysis was performed on SPSS 14.0 for Windows (SPSS, Inc., Chicago, IL). A p-value of <0.05 was regarded as statistically significant.

The faculties

The two faculties were full-time administrators from the Quality and Safety Division of the HA. A health care administrator was defined as a person who worked in a field relating to leadership, management, and administration of hospitals, hospital networks, and health care systems. Our two faculties were involved in policy-making, resources allocation, and system design concerning patient safety in all public hospitals in this locality. Both had over 2 years of experiences in incident management. One of the faculties had previously worked as an anesthesiologist and the other a staff nurse. They participated in our program on a voluntary basis.

The teaching program

The teaching program consisted of two 60-minute whole-class lectures designed to cover eight of the 11 topics listed in the WHO curriculum guide (World Health Organization 2009) (Table 1). The lectures used contemporary medical incidents in this locality as illustrative cases. These incidents had been handled personally by our faculties, who were thus able to present in-depth analysis and real-life management strategies. These cases were published in a bimonthly newsletter, which was required reading for our students and could be downloaded from a web-site (http://www.ha.org.hk/haho/ho/psrm/HARA14th.pdf). There was no formal assessment after the lectures.

Table 1.  Suggested topics listed in the WHO curriculum guide on patient safety

1. What is patient safety?^a

2. What is human factor and why is it important to patient safety?^a

3. Understanding systems and the impact of complexity on patient care.^a

4. Being an effective team player.^a

5. Understanding and learning from errors.^a

6. Understanding and managing clinical risks.^a

7. Introduction to quality improvement methods.^a

8. Engaging with patients and carers.

9. Minimizing infection through improved infection control.

10. Patient safety and invasive procedures.

11. Improving medication safety.^a

Note: ^aTopics covered in the teaching program of this study.

Results

A total of 100 out of 130 students completed the pre- and post-tests, yielding a response rate of 83.3%. Thirteen of the 23 items (56.5%) showed statistically significant changes at the post-test; 10 (43.5%) remained unchanged (Tables 2 and 3). Of the 13 items with significant changes, all were in the directions intended by our faculties.

Table 2.  Questionnaire items which exhibited significant changes at 3 months after the two hourly lectures on patient safety

	Before teaching		At 3 months after teaching
Items	Mean	SD	Mean	SD	Direction of intended changes	Detected changes	p-value
Nature and handling of medical errors
1. Human errors in medicine is inevitable	3.62	0.966	4.06	0.763	+	+0.44	0.001
2. After an error occurs, the most effective strategy is to work harder and to be more careful	3.44	0.931	3.05	1.058	–	−0.39	0.003
3. Competent physicians do not make medical errors	2.00	0.605	1.86	0.603	–	−0.14	0.030
4. The traditional culture of medicine is effective in dealing with errors constructively	2.81	0.798	2.60	0.778	–	−0.21	0.013
5. If you knew another physician has made an error, you would keep it to yourself	2.53	0.716	2.32	0.667	–	−0.21	0.009
6. If there is no harm to a patient, there is no need to address an error	2.33	0.750	2.16	0.666	–	−0.17	0.011
7. Disciplinary actions for staff are effective in preventing errors	3.20	0.794	2.60	0.991	–	−0.60	<0.001
8. There is no need to openly disclose an error if the patient has accepted an apology and is adequately compensated	2.77	0.973	2.48	0.904	–	−0.29	0.004
Self-appraisal of knowledge
1. The number of reported incidents each year in HAHO	2.06	0.755	2.55	0.803	NA	+0.49	<0.001
2. The causes of medical error	2.55	0.769	2.93	0.708	NA	+0.38	<0.001
3. Definition of latent factors	2.11	0.792	2.79	0.846	NA	+0.68	<0.001
4. Approaches to manage incidents	2.27	0.764	2.93	0.809	NA	+0.66	<0.001
5. Content of the publication ‘Risk alert’	2.24	0.872	2.93	0.887	NA	+0.69	<0.001

Note: NA, not applicable.

Table 3.  Questionnaire items which did not exhibit significant changes at 3 months after the two hourly lectures on patient safety

	Before teaching		At 3 months after teaching
Items	Mean	SD	Mean	SD	Direction of intended changes	Detected changes	p-value
Nature and handling of medical errors
1. A reporting system helps to reduce future errors	3.98	0.604	3.94	0.769	+	−0.04	0.318
2. There is a gap between what is know as ‘best care’ and what is being provided on a day to day basis	3.96	0.654	3.92	0.598	+	−0.04	0.724
3. Physicians should not tolerate uncertainty in patient care	2.74	0.896	2.75	0.892	–	+0.01	0.875
4. If you have made an error, you would not disclose it unless absolutely necessary	2.46	0.788	2.36	0.677	–	−0.10	0.193
5. An open reporting system has a negative effect on your professional status	2.58	0.812	2.53	0.826	–	−0.05	0.414
6. Most errors are caused by physicians or nurses	3.10	0.822	2.92	0.748	–	−0.08	0.253
7. Physicians are the best people to determine the causes of a medical error	3.00	0.766	2.96	0.777	–	−0.04	0.264
Patient safety in medical curriculum
1. Patient safety is an important topic	4.47	0.523	4.49	0.533	NA	−0.02	0.194
2. Patient safety should be included in the undergraduate curriculum	4.07	0.640	4.11	0.699	NA	+0.04	0.310
3. Professional examination should include items on patient safety issues	3.68	0.733	3.67	0.725	NA	−0.01	0.860

Note: NA, not applicable.

Responses to the 15 items concerning the nature and management of medical errors were screened for ‘intended changes’. Eight (53.3%) showed ‘intended changes’ (Items 1 to 8, Table 2). The greatest degree of improvement was in the students’ awareness of medical error being inevitable (Item 1, Table 2) and the notion that disciplinary actions and ‘being more careful’ may not be the most effective strategies for the prevention of error (Items 2 and 7, Table 2).

Seven of the 15 items (46.7%) concerning the nature and management of errors did not show significant changes (Items 14 to 20, Table 3). This group included two items to which students’ responses were mostly ‘positive’ at the pre-test and did not show further changes after the lectures (Items 14 and 15, Table 3). The remaining five items were identified as what were resistant to our educational intervention and represented areas for improvement. These included items concerning error reporting, uncertainty in medial care, and possible contributions made by non-clinical personnel in the causation and handling of errors (Items 16 to 20, Table 3).

All five ‘knowledge items’ showed significant intended changes but the mean scores of our students’ self-appraisals were all in the range of ‘fair’ or ‘poor’ only at both pre- and post-tests. This may suggest a better awareness of patient safety issues rather than any actual improvement in knowledge (Items 9 to 13, Table 2).

Our students regarded patient safety as an important topic at both pre- and post-tests. They were supportive of the teaching of patient safety and its inclusion in professional examinations (Items 21 to 23, Table 3).

Discussion

Patient safety has emerged as a distinct health care discipline which emphasizes incident management and risk reduction strategies (Willeumier 2004). The alarming number of patients reportedly harmed and killed by medical errors has led to the development of numerous trans-disciplinary and evidence-based strategies to improve safety (Brennan et al. 1991; Frush 2008; Graber 2009). These measures necessitated the conscious and dedicated participation from personnel at all levels. Medical students, as future health care providers and leaders, are now expected to be conversant and skilled in these areas. Reported studies, however, have identified deficiencies in patient safety training, and an explicit curriculum on patient safety is only available at a limited number of medical schools (Fulton 2004; Madigosky et al. 2006; Patey et al. 2007; Newell et al. 2008; Alper et al. 2009; Flin et al. 2009).

The implementation of patient safety teaching can be challenging. It involves a fundamental change in culture and the introduction of concepts outside traditional medical teaching such as risk management, system thinking, and quality improvement methodology (Willeumier 2004). Not all medical educators recognize its importance, and the introduction of a new program into an already over-loaded curriculum may be hindered by staff resistance, time constraint, and resource limitations. Moreover, each medical school also has its unique strength, limitations, and culture, to which a new program on patient safety must be tailored. The WHO curriculum guide on patient safety advocates the enhancement of existing parts of a curriculum, followed by the addition of new components at a later stage (World Health Organization 2009).

Both brief (Halbach & Sullivan 2005; Patey et al. 2007) and extended (Mayer et al. 2009) curricula on patient safety have been described, and the formats may vary from the use of team-based patient safety events (Anderson et al. 2009) to video-assisted simulation (Flanagan et al. 2004; Mayer et al. 2009). Improvement in students’ knowledge and awareness has been reported although not all positive changes were sustained (Halbach & Sullivan 2005; Madigosky et al. 2006). Our pre-test revealed certain deficiencies in patient safety culture amongst our students, and the information was useful for the planning of our program (Flin et al. 2009). We found that positive changes could be achieved even with a brief lecture-based program. It was also encouraging to know that our students were aware of the importance of this subject and were supportive of its introduction.

Finding the critical number of interested and informed staff to participate in patient safety education is another major difficulty. Patient safety is a relatively new subject and few faculties at our institution are familiar with its essential components. This is also likely to be the case at some other medical schools. Admittedly, medical staff are not the only, nor necessarily the best, people to teach patient safety. Nursing and pharmacy personnel, on the other hand, may be in better positions to introduce their existing safety improvement strategies. Psychologists may elucidate on the nature of human errors, and engineers can teach on system factors, safety culture, and human factors engineering. Our study indicates that administrative staff may also function as effective teachers. The term ‘health care administrator’ encompasses personnel with a wide range of backgrounds, expertise, and experiences. Our faculties have clinical backgrounds and we feel that our students can relate to them more readily. The faculties can also illustrate basic safety principles by discussing the management of contemporary and media-sensitive incidents which brings relevance and focus into their teaching.

Patient safety is a vast subject and our pilot program can only touch upon a few of the important topics. But we believe that it is better to start simple than not to have started at all; and even brief educational intervention has previously been shown to produce improvement (Paxton & Rubinfeld 2009). Our brief programs have failed to promote a healthy reporting culture amongst our students, and they continued to rate their knowledge in patient safety as ‘fair’ or ‘poor’ only. These are important areas for improvement and it is likely that a more comprehensive and longitudinal program is needed (Fischer et al. 2006; Patey et al. 2007; Pearson et al. 2010). We have since introduced patient safety in other clinical years in order to ensure the maintenance of new knowledge. We also align patient safety teaching with our students’ current leaning experiences by, for example, introducing the concept and use of an operating theatre checklist during their surgery clerkship. Whole-class lectures are by no means the best way to teach patient safety but we find it the easiest format to integrate into a busy curriculum (Alper et al. 2009). Clinical placement, small group tutorials, problem-based learning, simulation laboratories, and online activities are alternative delivery methods, all of which require significant resources input and meticulous remapping of an existing curriculum.

The main limitation of our study is the use of a non-standardized survey instrument that has not been validated for reliability or validity. We adapted it from a previous study and excluded some skill items as we did not plan to teach those skills in our pilot program. Standardized instruments for the assessment of patient safety culture are mainly catered for health care personnel such as clinicians and managers (Romano et al. 2009). A valid instrument targeted at medical students has recently been developed but unfortunately had not come to our attention earlier (Carruthers et al. 2009). The follow-up period in this study was short and we were only able to assess short-term impact. The very act of participating in the post-test may carry certain prompting effects. That, however, may be constructive in its own right.

Conclusion

We found that a brief program on patient safety may be readily integrated into an existing curriculum, and was able to produce at least short-term, positive changes. Health care administrators may be considered as effective faculties. The implementation of an explicit curriculum on patient safety can be demanding and should be tailored to individual institutions’ needs, strength, limitations, and culture. But even a simple program making use of available human resources from outside the medical school can be of value as a platform for future development.

Acknowledgments

The authors thank Dr Libby H.Y. Lee and Mr Fred Chan of the Quality and Safety Division, Hospital Authority Head Office (HAHO), Hong Kong SAR, China. They also thank Miss Gloria KB Ng, The University of Hong Kong for her assistance with data analysis and the preparation of this manuscript.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.