Interest in research among medical students.

While the only reliable method for probing interest in medical research is assessing actual voluntary research involvement, survey data (self-reported interest) may provide insights into any discrepancies between interest and actual involvement. To avoid pooling survey data that are too heterogeneous, we made a distinction between survey questions that ask about general interest in research and those specifically asking medical students about their interest in making commitments to research during their future careers.

I1a: Interest in performing research: A pooled weighed estimate of 72% of medical students reported having interest in performing research (0.72, 0.57–0.83). One particularly high estimate was that reported by De Olivera and colleagues, which showed that 90% of its 1004 student sample had interest in performing research [74]. However, even when this study was excluded from the analysis as a possible exception, the pooled weighed estimate remained fairly high (0.67, 0.53–0.79) (S4 File).

I1b: Interest in a career involving research: The single best estimator of career intentions of US medical graduates is probably the Graduation Questionnaire (GQ), developed by The Association of American Medical Colleges (AAMC) in 1978 [7]. In 2013, 63% of the 13,180 respondents indicated intentions to become somewhat-to-exclusively involved in research during their medical careers, including 17% who planned "significant" or "exclusive" future involvement. This huge sample size approaches a true census, with 81.8% of the US fresh medical graduate population being covered.

Upon quantitative pooling of our included studies, we found that about 31% of medical students (0.31, 0.19–0.46) were interested in a career involving research, and 12% (0.12, 0.07–0.21) showed interest in "significant" (higher) commitment to research during their future careers. One particularly important, high-quality study was that of McManus and colleagues, showing that 6.9% of UK medical students planned to pursue academic careers (or found them to be very appealing) [85]. When we calculated the pooled outcome excluding MacManus et al or the AAMC data, the pooled proportion was not markedly changed (S4 File).

It should be noted that there is considerable variation in the proportions reported in our included studies. This may reflect inherent (true) variability in students' research interests due to diversity of settings and study populations (as has been discussed in S2 File). We also believe that there are other potential contributors to this variability, most notably the ambiguity of wording of survey questions. For example, many studies did not make a clear distinction between interest in an academic (university faculty) medicine career, and interest in a career involving some research outside of academia.

I2. Medical students’ exposure to research.

Even today there is no consistent way in which undergraduate medical students are incorporated into research. For example, students may be engaged in research through summer research electives [9,45], mandatory curricular study modules [90], extracurricular research activities [93], or they might decide to intercalate for one or more years to obtain a BSc beside their medical degree. In Germany, it is mandatory for medical students to submit a thesis outlining the results of a research project in order to graduate with the title "Doctor" [30]. This requirement has also been reported in Peru, Finland, France and some U.S. universities such as Yale [24,27,76,94]. The AAMC 2013 Graduation Questionnaire shows that 68.2% of US medical graduates participated in a research project with a faculty member on a mandatory or volunteer basis and 41.7% co-authored a research paper [7].

If we exclude papers describing medical schools asking for mandatory graduation theses or research modules, we find that a little less than one third of medical students participated in research projects (0.31, 0.22–0.41). The proportion exposed to “prolonged” periods of research (>6 weeks) is even less (0.22, 0.16–0.28).

In the U.S., different medical schools have different research expectations, and the exposure of medical students to non-mandatory research seems to be largely dependent on medical school influence. Duke University, for example, incorporates students into summer-long research projects [95]. On the other hand, Stanford University, the University of Pittsburg and Warren Alpert Medical Schools incorporate students into longitudinal research projects in parallel with their academic studies [95–97]. This longitudinal approach may help in solving some of the reported problems of time-out research, such as the reluctance of medical students towards detachment from their colleagues and financial worries about spending extra time in college. Indeed, the success of Stanford is particularly evident, with 90% of medical students participating in research projects [91].

We found that the pooled proportion of medical students reporting some interest in research is higher than that of students who were actually involved in research projects. This may be due to: a) self-reported interest may not necessarily reflect serious willingness to pursue research; or b) lack of opportunities to meet students’ interest due to lack of funding, supervision and encouragement or inflexible curricula that leave little or no time for research (S5 File) [45,47–50,52,55,57,68,74].

II1. Effect of previous research experience.

Students who participated in research projects during medical school were over three times as likely to report interest in research involvement during their future careers (OR = 3.55, 1.84–6.83). Two studies [92,98], which were not included in the pooled weighed estimate, reported paired outcomes, with non-significant differences in research career interests after research exposure. Additionally, we found that medical school research involvement has no significant correlation with attitudes or motivation towards research (OR = 2.05, 0.99–4.24).

It is difficult to conclude that self-reported interest is a direct effect of exposure to research, since reverse causality cannot be excluded. That is, it is logical to assume that a fairly large proportion of students who had pre-existing interest in a career in research decide to participate in research projects. As a matter of fact, students in two of the included studies agreed that research participation strengthened pre-existing interest in a research career [90,91]. These findings also make sense in light of the fact that over half of all medical students reported having some interest in a career involving research (Fig 3). Another possible explanation for the above results is that students who have had prior research experience have better research knowledge and skills, and are therefore more confident about their ability to succeed were they to undertake research projects during their future careers. Indeed, in a series of interviews conducted by Jones et al, students who undertook an intercalated BSc in primary healthcare reported a positive influence of the experience on their appreciation of the research process [99]. Similarly, a thematic analysis of 905 SSC (Student Selected Component) projects by Murdoch-Eaton et al provided by medical students at six UK medical schools revealed gain of various research-related skills [90]. These results are also supported by eleven quantitative studies, summarized in S5 File [11,37,39,40,46,47,55,64,82,89,91].

II2. Effect of having a higher degree (MSc or PhD) prior to medical school.

II2a: Having a higher degree is associated with involvement in- (or planned involvement in-) research: Siemens et al report that medical students who had a higher degree prior to enrolment in medical school were almost four times more likely to perform research during medical school (OR = 3.95, 2.22–7.01) [52]. However, data provided by Cruser et al showed no significant difference between the two groups regarding their planned involvement in future research (OR = 1.01, 0.57–1.79) and Gerrard et al actually reported the reverse trend, with higher degree graduate-entry medical students actually being less likely to pursue an iBSc [54,81]. This is consistent with data we obtained from Mahesan et al, which shows that graduate-entry medical students (having any degree prior to matriculation) were almost ten times less likely to pursue an intercalated degree (OR = 0.01, 0.00–0.13) [62].

Since career progress (especially the pursuit of competitive residency) is a major motive behind medical student research, it may be argued that medical students with a higher degree view this aspect of their Curriculum Vitae (CV) as being “complete enough” and hence devalue the pursuit of another degree. In fact, to the medical student with a prior degree, an iBSc will almost always result in degree duplication, even if the skills and knowledge base of the iBSc course were completely different from those of the other degree already gained by the student.

II2b: Other advantages of having a higher degree (MSc. or PhD.): There is no significant correlation between having a higher degree prior to medical school enrolment and research interest or motivation. However, as might be expected, higher degree graduate-entry medical students were more knowledgeable about research, showed better research skills and had higher confidence in their research competencies (Fig 4). This is expected, given that almost all higher degrees have a compulsory research component.

II3. Effect of academic success.

II3a: Academic success is associated with attitudes towards basic medical sciences or medical research: The data we obtained from Hren et al shows an association between higher Grade Point Average (GPA) and attitudes towards research (OR = 1.83, 1.42–2.36) [79]. Cruser et al’s data, on the other hand, shows no significant difference between highest MCAT (Medical College Admission Test) scores and attitude scores [54]. Perhaps GPA during medical school, but not before admission, is a factor that influences attitudes. However, we believe the evidence in favor or against this hypothesis is weak and further investigation is needed in the future.

II3b: Academic success is associated with involvement in- (or planned involvement in-) research: The weighed pooled odds ratio from four included studies shows no association between academic success and involvement (or planned involvement) in research projects (OR = 1.00, 0.62–1.64). The only study showing a significant correlation was Brancati et al, which asserts that students who were academically successful (top third of their class) were more likely to choose an academic career (OR = 2.11, 1.30–3.42) compared to their less successful peers (lower third) [32]. However, this study investigates choice of an academic career rather than involvement (or planned involvement) in research during or right after medical school. Hence, it may be argued that this study should be excluded from the analysis as it measures a different outcome, in which case the pooled odds ratio remains non-significant (0.82, 0.59–1.15). We suggest further investigation into this issue using studies with more favorable, preferably prospective, designs.

II4. Financial factors affect the appeal of research to medical students.

About half of medical students who chose not to get involved in research reported being deterred by financial factors (0.50, 0.46–0.54) (Fig 5) [55,57,59,67,82]. Nicholson et al and Stubbs et al both show that about half of medical students who choose not to intercalate do so for financial reasons [59,82]. In addition, Galletly et al also reported that about half (48%) of medical students asserted that perceived lower salaries of academicians was an important factor behind their decision not to pursue an academic career [55]. The consistency of the findings by the former two studies with the latter one suggests that it's not just the short-term financial burden of pursuing an intercalated degree that deters medical students from getting involved in research, but a general long-term financial concern. Financial worries, particularly the fear of running out of grant money and the financial stress of academic careers, were indeed cited by students interviewed by O'Sullivan et al among the deterrents to academic career pursuit [100].

Similarly, Yamazaki et al and Kumar et al both showed that a considerable fraction of the general medical student population displayed concerns about the financial stability of a research career (45% and 12%, respectively) [57,67].

II5. Career progression is a main motive behind performing research during medical school.

The result from seven included studies indicate that career progression is a main motive (if not the main motive) behind performing research during medical school. These results indicate that in a large fraction of cases, medical students perform research for purely pragmatic reasons (related to their residencies or further post-graduate education), rather than pursuing research for the value it has in and of itself (Table 1) [48,49,52,54,55,82,86].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Career progression (eg securing competitive residency) is a main motive for medical students to perform research.

https://doi.org/10.1371/journal.pone.0127470.t001

Four studies mentioned the role competitive residencies play in driving medical students to perform research, and in fact students in three of those studies believed that seeking competitive residency was–explicitly- the main motive to perform research during medical school. The results from a qualitative study by Shapiro et al support this conclusion by showing that the motives behind research participation include (but are not limited to) pragmatic targets such as improving the students' relationship with faculty [101].

These conclusions are consistent with other results reported here showing that: a) there is a discrepancy between interest in clinical practice and interest in a research career (S5 File) [45,51,56,57] and b) there is a correlation between interest in academia or basic medical sciences and interest in research (S5 File) [55–57].

Combined, these findings indicate that any policies aimed at boosting medical students’ engagement in research have to align research involvement with the career progress and success of students. In much the same way that peer-reviewed publications are a key competitive edge in academia and in competitive residency applications, it must become clear that research is more than just an accessory when it comes to ordinary clinical practice.

II6. Other factors related- to or affecting medical student research.

As Reynolds has discussed, it is simply not enough to match students with professors in research projects, as good quality research requires real mentorship [102]. Research instructors also act as role models to encourage students to pursue careers in academic medicine. Further, finding the right mentor is important to ensure that students provide a working and intellectual input into the research projects, rather than simple assistantship in lab work or data collection (Table 2) [48,52,57,58,82].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Positive effect of mentorship or the presence of an academic role model.

https://doi.org/10.1371/journal.pone.0127470.t002

This is not always going to be easy; the results from two qualitative studies show that the complexity of ethical approval procedures (whether in terms of time or paperwork) is a major difficulty facing supervisors and students alike [90,103]. Further, the absence of clear, well-structured research governance may result in some aversion to faculty-mentored student research. This was the case in two qualitative studies, where students cited problems with approachability of faculty members and expressed concerns about being used as "free labor" on research projects [90,101].

In fact, Murdoch-Eaton et al's aforementioned project content analysis, while revealing some gain in useful research skills, also highlighted the failed attainment of a balanced skill-set; the majority of student projects involved information gathering and data processing, while fewer projects involved actual student engagement in research methodology development or critical analysis of data [90].

It may be presumed that the relatively short duration of the undergraduate research experience could limit its publication or citation potential. Indeed, Dyrbye et al found that graduates with a 17–18 week-long research experience published significantly less papers in which they appeared as first authors than their peers who spent 21-weeks doing research [29]. Further, Fede et al showed that the annual Undergraduate Medical Congress of ABC foundation (COMUABC) had a smaller proportion of abstracts accepted for publication in peer-reviewed journals in comparison to conferences of practicing physicians [70]. Conversely, Van Eyk et al. reported that the average number of citations of Dutch medical student publications was actually higher than the average citations for papers in the same field. [41]

A number of studies investigated factors that prevent medical students from being involved in research. Poor mentorship, lack of role models and perceived lower salaries of academic physicians were among the key factors cited (S5 File). The previous findings were also supported by four qualitative studies (Table 3) [17,45,90,99–101,103,104].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Studies with a qualita3tive component.

https://doi.org/10.1371/journal.pone.0127470.t003

In addition, institutional influence as well as the type and length of available research opportunities were found to be relevant factors in determining whether students choose to engage in research [51,53]. McLean and co-authors provided an excellent set of tips to bolster the involvement of students in academic medicine projects and potentially overcome some the aforementioned limitations [105].

. The importance of psycho-cognitive factors in determining medical students' motivation towards- and engagement in- research was also highlighted in the qualitative literature. One of the most important motives behind performing research is curiosity. Not only is curiosity a main motive behind pursuing research while in medical school (as has been shown by Shapiro et al [101]), it is one of the very early psycho-cognitive predictors of persistence into scientific or research disciplines even before enrolment into medical school [17,104]. Conversely, perceived lack of competence may deter medical students from pursuing research-active careers [45].

III1. Medical student research results in a publishable product.

Peer-reviewed journal publications are generally considered to be the best indicator of research productivity, and it may be viewed as a major metric (though not the only one) of the “return on investment” in supporting and funding medical student research. An average of 30% (0.30, 0.19–0.44) of research performed by medical students resulted in a peer-reviewed journal publication. When only higher quality publications were included in the analysis (indexed in Medline, Scopus or Medic), the proportion remained more or less the same (0.31, 0.18–0.47). Subgroup analysis of studies investigating the research productivity of graduation theses revealed that 26% (0.26, 0.10–0.52) of graduation theses result in higher quality publications.

As expected, all studies reporting first-authored peer-reviewed publication by medical students described instances of prolonged research exposure. An average of 13% (0.13, 0.06–0.27) of medical student research resulted in a first-authored peer-reviewed publication. The pooled outcome remained the same when only higher quality publications (Medline-, Scopus- or Medic- indexed) were included in the analysis (0.13, 0.05–0.30).

A few initiatives, aimed at propping up medical student publication output, have gained popularity over the last few years. Those initiatives include a number of student-run journals and journal spaces dedicated solely for medical student research publications [107–110]. A subset of these journals is Medline-indexed and some even involve undergraduates in the peer-review process. Similarly, the Yale Journal of Biology and Medicine annually publishes Yale's student thesis abstracts [111]. These initiatives, we suppose, will help in promoting student participation in research and comfort students about publication issues. To our knowledge, there is no systematic investigation in the literature so far regarding the quality of research published in medical student research journals in comparison to field-specific journals. Hence, we would like to take a conservative stance whenever we see such hierarchical "segmentation" of the scientific enterprise; the stringency of research assessment, in our opinion, should be indiscriminant to the identity of the study authors.

It is important to note that the failure of publication of medical student research may be reflective of other factors beside the success and relative contribution of the student. For example, Weber et al showed that 55% of the papers submitted to a medical specialty conference did not reach the stage of publication five years later [112]. Similarly, Riveros et al found that half of the clinical trials reporting results in ClinicalTrials.gov had no corresponding journal publication [113]. Keeping this in mind, the results by Cursiefen et al should not be surprising; showing that medical students were among the authors of 28% of the papers produced by a German medical faculty, even though only 66% of medical student research resulted in a publication [30].

III2. Research during medical school is associated with later involvement in research projects.

Students who took part in research projects during medical school were more likely to get involved in (or report planned involvement in-) research later in their careers (OR = 3.58, 1.82–7.04). When a subgroup analysis was performed to include only studies that explicitly refer to academic careers (as opposed to brief research encounters), students who performed research during medical school were over six times as likely to pursue academic careers (OR = 6.42, 1.37–29.98) than their “untreated” peers.

With one exception, none of the included studies had a prospective design; hence reverse causality cannot be excluded, and is in fact very likely (students planning academic medicine careers choosing to get involved in research during medical school). Indeed, the only prospective study included (McManus et al [85]) showed that at the time of application to medical school, students who later chose to take an intercalated degree were already significantly more likely to report definite or highly likely choice of academic medicine careers (OR = 1.37, 1.13–1.66). Just before graduation, however, this likelihood had a substantial increase (OR = 3.45, 2.27–5.24). Together, these results indicate that medical school research strengthens pre-existing interest in an academic career.

A qualitative study by O'sullivan et al emphasized the value of early research exposure in giving medical students the opportunity to entertain the thought of pursuing academic careers [114]. Such exposure, they concluded, may sometimes even discourage students from pursuing academia, but is necessary nonetheless given the lack of sufficient free time during post-graduation residency to experience research.

III3. Research during medical school is associated with long-term success in academia.

Three studies showed that physicians who performed research during medical school were more likely to attain faculty rank long after graduation [8,32,66]. While this has implications on the decision of individual medical students to pursue research, we argue that it has little bearing on policy decision-making, since faculty positions are awarded on a competitive basis. Indeed, Brancati et al showed that this effect was dependent on the publication status of research performed during medical school [32]. In other words, students who did not publish their research were not significantly more likely to attain higher faculty rank on the long run. Hence, the fact that medical student research is associated with higher likelihood of attaining faculty positions has little implications regarding the systematic incorporation of research into medical curricula.

Students who performed research during medical school were more than twice as likely to author at least one peer-reviewed publication later in their career (OR = 2.31, 1.88–2.83). This remained true after the exclusion of Chusid et al [25] (which correlates successful publication of graduation theses with long-term publication success) from the analysis (OR = 2.26, 1.83–2.77). They were also twice as likely to acquire first-authorship (OR = 2.21, 1.56–3.13). The total number of publications and ability to secure grants, too, was reported to be significantly higher among students with medical school research experience [81]. Evered et al, on the other hand, found no significant difference in either of those measures between both groups [66]. Moreover, students who performed research during medical school were more likely to be cited at least once [66], had a higher total citation count [81], were more likely to be cited more than 20 times [66], and had higher odds of receiving awards [8,81] later in their careers.

While this data provides strong evidence of a correlation between medical school research and long-term success in academia, a causal relationship cannot be established since students who decide to perform research may already have a keen interest in research. Nonetheless, a causal relationship is quite likely since early research experience (especially if it culminates in a first-authored publication) would naturally enhance the career prospects and significantly improve the CV’s of early career medical graduates. Overall, we believe that the long-term impact of medical school research is inadequately assessed, and that further evidence is needed using prospective study designs with proper adjustment for baseline status.

III4. Research during medical school is correlated with career choice of- (or interest in a career in-) the same or related specialty as the research project.

Three of the studies that met the broad inclusion criteria reported results from control or “untreated” groups. Other studies reported results only from treated groups and hence were excluded from the analysis. Overall, students are 2.7 times as likely to be interested in careers in the same (or related) clinical specialty as the research project they got involved in during medical school. As with many other conclusions in this review, a causal relationship cannot be determined from this apparent correlation. This is especially true in the case of competitive residencies (and is particularly relevant to US residencies), where research experience in the same specialty gives recent graduates a competitive edge over their peers without such experience.

The relationship between medical school research and clinical practice was also touched upon in two of the included qualitative studies. Shapiro et al showed that many faculty members mentored student research in family practice in order to attract students to the same specialty [101]. Indeed, students interviewed by Jones et al believed an iBSc in primary healthcare provided them with deeper insights into patient care and a more thorough understanding of evidence-based clinical practice [99].

IV1. Effect of prolonged research time-off on subsequent academic performance.

One of the issues discussed in the literature is the effect of prolonged research time-off amid the medical program on subsequent clinical knowledge. This question has been assessed in the context of iBSc degrees in a recent review [119]. All but one of our included studies investigated the effect of taking an intercalated degree on subsequent academic performance. The results have been conflicting; two studies that either matched groups by previous performance or adjusted for pre-clinical scores found no evidence of improvement in scores [88,120]. All five other studies that met our inclusion criteria reported an improvement in academic performance.

Due to heterogeneity in academic assessment methods and high possibility of confounding, we only pooled the studies for which we could extract odds ratio values that: a) measure final year academic scores and b) control for previous academic performance. Three studies met these two inclusion criteria, all of which reported the effect of iBSc degrees. On average, students who took some time off to perform research were twice as likely to outperform their peers (OR = 1.99, 1.39–2.84), even after adjustment for previous academic performance. It is noteworthy that all pooled studies investigated research time-off that was around one year in duration (iBSc), and that the positive effect of research time-off on subsequent academic performance may actually be reversed if the research delays are prolonged. Dyrbye et al pinned down a critical period of three years, after which medical students start to lose clinical knowledge and skills by the time they return to the core medical program [28].

IV2. Gender equality in medical student research.

There is no apparent gender difference regarding the following outcomes: Interest in a career in research; involvement in research during medical school; attitudes towards research; interest in- or motivation towards- performing research; research knowledge or skills. However, on average, males seem to be significantly more likely to publish (or submit for publication) the research they performed during medical school (OR = 1.59, 1.26–2.01). The reasons behind this gender gap in publication are unclear to us, and have been inadequately researched. Since there is no apparent gender difference in research perceptions, attitudes, motivations or knowledge, we suspect that the gender difference in publications is due to factors unrelated to research such as the overall academic environment or psychosocial factors. Indeed, these findings are consistent with a 2006 study by Jagsi et al showing a generalized gender gap in the authorship of academic medical articles in six major medical journals. Whatever the reasons behind gender differences in publication, they underlie a general issue not specific to medical school research [121].

IV3. Type and field of research performed by medical students.

The majority of medical student research is original in nature (as opposed to literature reviews). We were interested in finding out what percentage of these research projects were in the basic sciences, since this issue is of particular relevance to translational research. We found that the proportion was highly variable between different studies. In four of the five included studies less than half of medical student research was lab-based basic research, and the pooled weighted estimate was 0.32, 0.14–0.49. Given the relevance of research to competitive residency applications, it should not come as a surprise that lab-based projects do not constitute the majority of medical student research. Nonetheless, these results indicate that efforts directed at increasing the number of physician scientists involved in translational research should not only be directed at bolstering research involvement, but also improving the appeal of basic lab-based research to medical students.

IV4. Compulsory vs. elective medical school research.

The question of whether undergraduate medical research should be made compulsory or elective has been discussed in the literature, and is a matter of debate [37,97,122]. Arguments in favor of mandatory incorporation revolve around the ever-increasing importance of evidence-based clinical practice, while arguments against it revolve around the importance of focusing on clinical skills education. Diez et al. recommended against Germany's dissertation requirement, due to the steady decline in the number of successful dissertations [123]. Our results tell a similar story; the fraction of graduation theses resulting in a first-authored higher quality publication was smaller than the overall average (0.07, 0.03–0.14). At first, this may seem counterintuitive, as one may predict that the systematic incorporation of research as a necessary graduation requirement would raise the fraction culminating in a first-authored higher quality publication. However, one needs to bear in mind that since graduation theses are an obligatory requirement, a fraction of those students performing research may not be interested at all in what they are doing. Taking this into consideration, it should not come as a surprise that percentages as high as 34% (Cohen et al [38]) and 31% (Dyrbye et al [29]) of voluntary medical student research were reported to result in first-authored Medline-indexed publications. Weihrauch et al and Pabst et al, on the other hand, reported favorable results in terms of the personal and professional value of the German dissertation requirement [124,125].

IV5. The situation in countries with developing economies.

We retrieved studies that were performed in India [67,72], Uganda [68], China [69], Brazil [70,74], UAE [71], Croatia [73,79], Pakistan [75,77,80,86], Peru [76], and Turkey [78].

The number of medical schools and the research budget in developing countries are alarmingly mismatched with their needs [1]. This disparity, we believe, reflects naturally on the status of medical student research. In fact, medical student research might be even more important in developing countries than in developed countries, due to the pressing need to adapt international standards to local community needs.

Medical students in developing countries arguably face a set of extra challenges and are influenced by a number of different factors in comparison to developed countries [126]. For example, the high student-to-teacher ratio makes it increasingly difficult for medical students to have mentors and role models. Even research based on statistical analysis of patient records is often difficult to perform in many medical schools, due to suboptimal Information and Communications Technology (ICT) infrastructure in hospitals and in teaching premises in countries with developing economies [127]. While excellent research may of course be performed in resource-poor countries, it is preferable that any reform in research funding is coupled with a well-developed educational and managerial infrastructure; otherwise the research output may be largely suboptimal [128]. Worryingly, an essay by Silva et al. reported a decrease in the ratio of Undergraduate Student Research Assistant Programs (USRA's) to the number of undergraduates in Brazil over the past years [129].

Students’ interest in research was higher in countries with developing economies than in developed countries (0.82, 0.67–0.91 vs. 0.47, 0.26–0.69). One possible explanation for this finding is that the lack of opportunities causes higher eagerness to perform research. Another, possibly more likely, explanation is higher career-related anxiety in lower-income settings, with a resultant boost in research interest. Indeed, students in developing countries were not significantly less exposed to research, a result which may be reflective of the higher interest rates, bolstering research engagement despite inadequacies in resources. These results are supported by the findings of Baig et al, showing that 40% of Pakistani medical students viewed research as a tool to secure competitive residencies in the US [86].