Evaluation of a clinical attachment in Primary Health Care as a component of undergraduate medical education

Abstract

Introduction: It seems that there is a trend in undergraduate medical education towards including clinical attachments in primary health care (PHC) worldwide. The benefits of such initiatives are already well described. The aim of this study was to evaluate the effect of a clinical attachment in undergraduate medical students in a European country with an odd medical educational system that essentially lacks any kind of academic PHC departments.

Methods: The study was undertaken during 2005–2007. A non-mandatory 1-week clinical attachment in PHC/general practice was organized in an urban PHC unit by general practitioners with educational experience in collaboration with the Department of Physiology of the local medical school. The participants were a prospective cohort of medical students in the 2nd year of undergraduate studies. All participating students sat a pre-defined clinical exam which consisted of multiple choice questions, mini case papers and an objective-structured clinical examination before and after the attachment. In addition, the students rated the whole process.

Results: The response ratio was 77.06%. The mean score on objective structured clinical examination of participants increased from 30.70/100 to 62.28/100 (p < 0.001). The students’ impression of the study was rather positive (4.39/5).

Discussion: The educational intervention of including a clinical attachment in an undergraduate curriculum seems to have encouraging results, considering the peculiarity of inexistence of academic departments of PHC or General Practice in the national medical schools and the inexperience of students regarding similar concepts.

Introduction

It seems that in recent years, undergraduate medical education has changed in terms of including attachments in primary health care (PHC) (Murray et al. 1997b; Howe 2001; Oswald et al. 2001). The aforementioned attachments in community hospital-based primary care, have confirmed their educational benefits and their popularity with both students and tutors (Grant et al. 1997). In the United Kingdom, there is a considerable number of medical schools which have already adopted community-based teaching methods (practices) of primary care clinical skills. In addition, a similar trend is delineated, with the transition and gradual shift from tertiary level hospital educational environments towards smaller and patient-oriented community PHC units (Robinson et al. 1994; Irby 1995; Murray et al. 1997a; Oswald et al. 2001).

Table

Practice points

Clinical attachments in PHC have already proven their effectiveness in countries with a well–established system of undergraduate medical education in general practice/PHC. Despite the lack of experience in such projects, the two-year experience of introducing a clinical attachment in a PHC unit in the curriculum of a medical school of a country that literally lacks any kind of PHC–oriented medical education helped the students improve their clinical skills. The contact of medical students for the first and only time with PHC during this attachment may have influenced them in terms of selecting general practice as a career choice after their graduation.

This phenomenon can be attributed, at least in part to the fact that tertiary level hospitals have a clearly different strategy from PHC units. Hospitals mainly focus on same-day treatments (one day clinic) and early discharge of patients, with only severe cases requiring longer hospitalization periods (Sturmberg et al. 2001; Bryant et al. 2003). The majority of patients seeking medical aid (either emergencies and acute conditions or cases with chronic diseases) are mainly served by PHC units and primary care physicians are facing the challenge of effectively and efficiently serving those patients. The necessary environment, offering the experience and knowledge for successfully acquiring the skills for this completely different approach to providing medical care, is that of a PHC unit (Nieman et al. 2006). As a result, undergraduate medical students who do not include community-based attachments in their curriculum miss out on important aspects of this particular category of health care service and medical education.

As already documented in literature, the benefits of placements in the community include the opportunity for medical students to consolidate their knowledge of longitudinal care (Worley & Lines 1999), of preventive medicine, of health promotion, and of the interaction between physician and the patients–all of the above being principal components of PHC. On the other hand, there is concern regarding both the quantity and the variety of clinical conditions a medical student will gain exposure to, especially if compared to a tertiary level clinic or emergency room (Howe & Ives 2001; Denz-Penhey et al. 2004), but at this point a discussion should be not be made for the purpose of favoring the one versus the other, but rather for incorporating and blending both in the students’ educational curricula.

It should be stressed that, although there is an increasing need for general practitioners worldwide, few medical students are choosing this medical specialty in Europe and North America (Kahn et al. 1999; Skinner & Newton 1999; Wright et al. 2004; Mariolis et al. 2007). Another benefit of community-based education is the increased exposure to general practice and family medicine, which seems to have a positive influence on medical students choosing this specialty as a career path (Furnham 1986; Morrison & Murray 1996).

As described in other studies, the results of community-based attachments in undergraduate medical education are surprisingly encouraging, sometimes surpassing their tertiary hospital-based counterparts in terms of improving the clinical skills of the students (Satran et al. 1993; Murray et al. 1997a).

Given that we live and practice in a country where general practice and family medicine are still not a part of the academic community (Mariolis et al. 2008), our experience may seem inadequate. Except for one academic department of social medicine, which has undertaken several initiatives to promote undergraduate and postgraduate education and training in general practice (Lionis et al. 2004), no other university hosts a general practice/family medicine department. Although there is a large need for general practitioners in our country, the majority of medical students opt for careers in other medical areas. Currently, the proportion of medical students oriented versus GP for a specialization is as low as 3.1%. Since 1986, when GP was established as a distinct medical specialty with an individual training program, the attention of policy makers has been focused on how to attract new medical graduates in order to attain a satisfactory number of primary care physicians, capable of providing the desired PHC services for the country's rural, and recently urban, population (Mihas et al. 2006). However, the focus of policy on improving the training of general practitioners and on providing incentives to make this career choice attractive, did not extend sufficiently to encourage the creation of departments of GP in medical schools and the inclusion of general practice or PHC in the medical curriculum (Mariolis et al. 2007), but was limited instead to measures designed in order to attract physicians in choosing GP as a career option by providing the conditions for an effortless and straightforward achievement of a permanent working position in a public community health center.

In an attempt to fill this educational gap and to test the hypothesis of a more effective and extended educational program, the Department of Physiology of the Medical School at the University of Athens (the Greek capital) in collaboration with the single urban PHC unit in Greece (Health Center of Vyronas), conducted an experimental, theoretical and practical attachment in general practice, in the first and second term of the second year of undergraduate studies in medicine. The aim of this study was to evaluate the impact of this attachment on medical students in terms of improving their clinical skills, as well as assess the effects of introducing general practice to them and their perceptions regarding this project.

Methods

Setting and participants

The study was undertaken during the academic years 2005–2006 and 2006–2007. The participants were a prospective cohort of medical students in the 2nd year of undergraduate studies at the local medical school. The school has a medical curriculum made up of three years of basic sciences followed by three years of clinical medicine. All students who attended the compulsory clinical physiology course as part of their curriculum (376 during 2005–2006 and 291 during 2006–2007) were invited to participate voluntarily in the clinical attachment upon conclusion of their regular daily schedule (i.e. after 3:00 p.m.) at the beginning of each academic year. After the application forms were collected, the students were randomly (adjusting for age and sex) allocated to groups of 15.

Each group started on a different week during the academic year. The clinical attachment and the study itself took place at our PHC unit from 16:00 to 21:00 Monday to Friday, and lasted for one week for each 15-student group.

Clinical attachment curriculum

The aims of the whole program were clearly set out in writing prior to its beginning. The primary goal was that all participating students should be able to acquire basic clinical skills of history-taking and clinical examination, understand the theory behind those skills, as well as practice their doctor-patient communication skills. The courses offered medical students a first-time contact with real patients and physical examination, along with exposure to the objectives and principles of medical physiology. The fact that the students were at the early stages of their education and lacked any clinical experience was also considered. The secondary goal of the program was to educate and familiarize students with general practice and family medicine, in order to redress the lack of a relevant compulsory undergraduate course.

The general practice tutors were all general practitioners (a coordinator and ten tutors), with each one having specialized in teaching a special subject. Each group of students was randomly divided into two subgroups at the beginning of each day. The participants were attached to one GP tutor for the first half of the training day, then to a different one for the remaining half, in order to compensate for any selection bias.

Measures and outcomes

Upon arrival at the Health Center on the first day of the attachment, the participating students were asked to fill in a questionnaire regarding demographic characteristics such as age, gender, if they had relatives in medicine, if they had ever won any distinctions, or if they also had a previous academic degree. Moreover, they were asked to rate how likely they were to opt for general practice as their chosen career on a five-point scale from ‘very likely’ to ‘very unlikely’.

In order to assess the students’ performance, all students sat a pre-defined clinical exam which consisted of multiple choice questions (20 questions, 40%), mini case papers (short report, 20%) and an objective-structured clinical examination (history-taking, physical examination, communication skills and data interpretation), which took place during encounters with standardized patients who represented common clinical findings (40-scale marking by the examiners, 40%). The exam produced a maximum score of 100. Furthermore, it was developed in a way to reflect common clinical problems in primary care and to evaluate the familiarity of the participants with common clinical skills, as well as their doctor-patient relationship. The examiners were two general practitioners who were not involved in any way in the educational procedure of this clinical attachment and were experienced in objective-structured clinical exams. In order to ensure the robustness of the assessment, double marking was performed in all participants. The mean score was finally used. The marking of the short reports and of the objective structured clinical examination was aided by guidelines that were constructed ad hoc. These guidelines indicated the approach that would be taken to marking, including attachment outcomes that are targeted by the question, criteria to be applied to responses along with the marks to be awarded in line with the quality of the responses and anticipated sample answers, covering a range of marks. The intraobserver and interobserver reliability were assessed using Spearman's correlation coefficient and Cronbach's alpha, respectively. For the purposes of the reliability testing, 25 participants (5% of total) were randomly selected and re-examined by the same observers, one day after their first examination.

On completing the attachment, students sat an exam similar to the beginning clinical exam. In addition, they repeated that part of the first-day questionnaire regarding general practice and were additionally invited to rate their preceptors, their level of satisfaction with the course, as well as quantify their sense of achievement in respect of their learning goals and expectations. The latter was performed using two five-point scales from ‘very poor’ to ‘outstanding’; and from ‘not at all’ to ‘very much’, respectively. The questionnaires were numbered in order to be linked by means of a referral list held by the coordinator of the study. All information given was treated in a confidential manner. Each preceptor received a report of the ratings given by his students.

The statistical department of the health centre kept records of the diagnoses made during the clinical attachment according to International Classification of Primary Care, 2nd edition (Wonca 1998).

Statistical analysis

Continuous variables are presented as mean values ±standard deviations, while categorical variables are presented as absolute and relative frequencies (percentages). The Shapiro–Wilk criterion was used for the assessment of normality. The student's t-test for paired samples was used for comparison of continuous variables while Wilcoxon sign–rank statistic was used to assess any differences in ordinal variables before and after the attachment. The Pearson's chi-square statistic was estimated for evaluating any association between categorical variables. All reported p-values are based on two-sided tests and compared to a significance level of 5%. Data were analysed using STATA statistical software (Version 9.0, Stata Corporation, College Station, TX 77845, USA).

Results

A total of 513 students (243 during 2005–2006 and 271 during 2006–2007) rotated through the non-compulsory attachment in the two study years. The response ratio increased from 64.63% in the first year to 93.13% in the second year of the attachment (243/376 and 271/291, respectively) All of the students who followed the attachment participated in the study. The basic characteristics of the study sample are described in Table 1. No statistically significant differences were found between the two genders in all demographic and educational characteristics. The mean correlation coefficients for intraobserver reliability were 0.89 and 0.91 (p < 0.001, before and after the attachment, respectively), while Cronbach's Alpha values for interobserver reliability were 0.84 and 0.87 (before and after the attachment, respectively).

Table 1.  Basic characteristics of medical students participating to the general practice clinical attachment

Variable	Men (N = 246)	Women (N = 267)	Total (N = 513)	p-value
Age	20.07 ± 1.12	19.98 ± 1.14	20.02 ± 1.13	0.368
Area of origin
Urban	119 (48.37)	134 (50.19)	273 (53.22)	0.427
Semi-rural	59 (23.98)	72 (26.97)	127 (24.76)
Rural	68 (27.64)	61 (22.85)	113 (22.03)
Knowledge of at least one foreign language (CEFR* level: C2)	29 (11.79)	33 (12.35)	62 (12.09)	0.843
Having at least one parent medical doctor	85 (25.68)	82 (23.50)	167 (24.56)	0.508
Having awarded at least one academic distinction	17 (6.91)	19 (7.12)	36 (7.02)	0.927
Having other academic degree	7 (2.85)	6 (2.25)	15 (2.53)	0.667

*Common European Framework of Reference for Languages: Learning, Teaching, Assessment.

The results of the clinical attachment as an educational intervention regarding clinical skills and patient physical examination are presented in Table 2. A significant improvement in all components of the objective-structured clinical examination has been observed after the intervention, a fact that can be easily depicted from the results.

Table 2.  Mean scores on objective structured clinical examination of participants before and after the clinical attachment

Skill	Before	After	p-value
History taking (maximum score 30)	11.28 ± 5.41	18.53 ± 4.34	<0.001
Physical examination (maximum score 30)	5.74 ± 1.89	16.83 ± 2.51	<0.001
Ability to interpret basic medical data (maximum score 30)	7.42 ± 3.63	19.01 ± 5.68	<0.001
Communication skills (maximum score 10)	6.26 ± 2.63	7.91 ± 1.84	<0.001
Total score (maximum score 100)	30.70 ± 3.24	62.28 ± 3.97	<0.001

Table 3 shows the positive effect of the clinical attachment on the possibility of selection of general practice as a career choice after graduation. The proportion of students who might choose to specialize in general practice almost doubled (18.71%, as opposed to 10.92% before the intervention). This increase was significant in both genders.

Table 3.  Probability of selection of general practice as a career choice before and after attachment

	Men (N = 246)		Women (N = 267)		Total (N = 513)
	Before attachment	After attachment	Before attachment	After attachment	Before attachment	After attachment
Likely/Very likely	31 (12.60)	53 (21.54)	25 (9.36)	43 (16.10)	56 (10.92)	96 (18.71)
Neutral	74 (30.08)	55 (22.36)	82 (30.71)	54 (20.22)	156 (30.41)	109 (21.25)
UnlikelyVery unlikely	141 (57.32)	138 (49.46)	160 (59.93)	170 (63.67)	301 (58.67)	308 (60.04)
p-value	0.014		0.004		<0.001

Looking at students’ ratings of their tutors during the attachment, as described in Table 4, their impressions could be characterized as positive judging by the mean overall rating being 4.51 (out of a maximum score of 5) while no preceptors’ attribute was graded below 3.97. In the same table, the overall impression of the whole procedure is also described.

Table 4.  Student's ratings and achievement of learning goals and expectations

Variable (maximum score 5)	Mean ± SD
Ratings of preceptors
Was experienced in his subject	4.13 ± 0.52
Was able to clearly respond to questions	4.24 ± 0.56
Showed patience	3.97 ± 0.84
Was enthusiast for his subject	4.53 ± 0.45
Gave opportunity to actively participate	4.33 ± 0.41
Was friendly	4.70 ± 0.28
Informed us about general practice	4.45 ± 0.33
Overall rating	4.51 ± 0.29
Own impressions
Achievement of learning goals	4.32 ± 0.54
Achievement of own expectations	4.15 ± 0.49
Overall impression of the attachment	4.39 ± 0.42

Discussion

The singularity of our study is the medical educational system of our country. No medical school in our country hosts an academic department dedicated to PHC and, as an inevitable consequence, no kind of undergraduate medical education (either as a theoretical course or a clinical attachment) related to this scientific subject is currently available to students. The primary aim of this study was to evaluate how a clinical attachment in a PHC setting benefits the development of medical students’ clinical skills during the second year of medical school, a fact which, although innovative within our academic reality, is not considered out of the ordinary for many other countries and institutions (Campos-Outcalt et al. 1995; Nieman et al. 2004). In addition, medical students were given the opportunity to have their first contact with clinical practice as seen within the framework of a PHC unit. The results suggest that students improved their clinical skills after the educational intervention and are in accordance with similar studies (Murray et al. 1997a; Wade et al. 1998). This is the first study which indicates that an introduction to clinical primary care via an early clinical skills attachment, can contribute to a significant improvement at the level of knowledge of those skills during early undergraduate studies in a medical school that lacks a primary care–general practice academic department and of course, any contact with this medical specialty. Such initiatives are already implemented (Carney et al. 1999) and it is clearly shown that community-based settings such as primary care offices–as in our case here–performed as well if not better than hospital-based environments.

A very encouraging fact which emerged from the results was that, not only did the students improve their clinical skills, but they literally embraced the whole procedure. It should not be forgotten that participation in this attachment was completely voluntary, and that the students also had to sacrifice a remarkable amount of personal time and practically exceed 13–14 hours in educational activities for a whole week period. The response ratio increased from 64.63% in the first year to 93.13% in the second year of the inclusion of the attachment in the curriculum, while the overall impression of the clinical attachment was very positive. This could be attributed to the fact that the students acquired clinical experience for the first time during this attachment and to the enthusiasm that may result from this. The variety of clinical conditions the students encountered might have also contributed to this result, given the ‘hospitalized’ way of medical thinking and educating that dominates the rest of the university curriculum. The achievement of the students’ own goals, which were set by themselves at the beginning of the study, are an added advantage of primary care clinical attachments (Nieman et al. 2006) as well as a potential reason for the students’ positive impression of them.

The general practitioners who volunteered their time to teach medical students, although experts in their own subject area, they lacked academic experience, for reasons detailed above. The high ratings they received from the students are certainly a strong incentive for them to continue to participate in this initiative. A secondary goal of this attachment was to familiarize the students with primary care and general practice in general. We think that the results rewarded our work, since a significant change in the likelihood of choosing this medical specialty as a career choice was recorded. This finding agrees with those of other studies on the same subject (Morrison & Murray 1996). It seems that this sort of initiative has a greater effect on medical students who have limited knowledge of primary care and how it is applied, as was the case with our sample.

Apart from all the obviously positive results, our study has some limitations. One limitation is the relatively small duration of the clinical attachment per student group (one week), which might have influenced the results. Unfortunately, given the elective nature of the attachment, a longer exposure for the benefit of the students was impossible. It is hoped that the encouraging results of this study could prompt a change in the curriculum of the Athens medical school, which could lead to the attachments increasing in duration. Another weakness of the study is that the intervention took place in only one PHC unit, a fact which clearly eliminates any possibility of performing a comparison between the Health Center of Vyronas and another PHC unit, or even better with a hospital setting in order to make available more interesting results. However, analogous studies have shown that medical students can learn clinical skills and perform equally well (if not even better) in a PHC-based teaching environment, as they would in a tertiary care setting (Murray et al. 1997a; Worley et al. 2004). It should be mentioned that there was no control group, not allowing us to adjust for environmental or other potential confounding factors that might have biased the results. Although the comparison with a control group would strengthen the results, we think that the observed differences before and after the attachment were large enough and fairly significant (or indicative) in order to suggest its benefit. In addition, there are numerous studies (Worley et al. 2004; Nieman et al. 2006) dealing with the same subject which did not actually use a comparison group. The change in students’ opinions regarding general practice as a career choice should also be validated after 4 to 6 years, when those students complete their medical studies and make their final selection. It is also possible that further changes may occur in the career choices of participating students and therefore they will be.

Conclusion

Although the clinical attachment evaluated in our study, does improve medical students’ clinical skills and familiarize them with general practice, it is still a drop in the ocean when one considers the curriculum of the medical schools in our country. Although this initiative began as an ‘educational experiment’, its results are quite encouraging and it might pave the way towards a primary care-oriented shift in the educational programs of Greek medical schools. Based on the international experience (Smith & Weaver 2006) and on our results, we have reasons to believe that the integration of similar attachments in the curricula of national medical schools will improve the quality of the studies.

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

Additional information

Notes on contributors

Anargiros Mariolis

ANARGIROS MARIOLIS, MD PhD is a General Practitioner and Head of Department of the Primary Health Care Centre of Vyronas.

Constantinos Mihas

CONSTANTINOS MIHAS, MD MSc and ALEVIZOS ALEVIZOS, MD are General Practitioners and associated with the Health Centre of Vyronas as clinical research investigators.

Alevizos Alevizos

CONSTANTINOS MIHAS, MD MSc and ALEVIZOS ALEVIZOS, MD are General Practitioners and associated with the Health Centre of Vyronas as clinical research investigators.

Marek Papathanasiou

MAREK PAPATHANASIOU, MD, and KONSTANTINOS MARAYIANNIS, MD MSc, are General Practitioners who work in the Health Centre of Vyronas.

Theodoros Mariolis-Sapsakos

THEODOROS MARIOLIS-SAPSAKOS, MD, is a General Surgeon and associated with the Health Centre of Vyronas as a clinical research investigator.

Konstantinos Marayiannis

MAREK PAPATHANASIOU, MD, and KONSTANTINOS MARAYIANNIS, MD MSc, are General Practitioners who work in the Health Centre of Vyronas.

Michael Koutsilieris

MICHAEL KOUTSILIERIS, MD, PhD, is professor and Head of the Department of Experimental Physiology, Medical School, University of Athens, Athens, Greece.