Background

Patient centred care occurs when healthcare providers are respectful of and responsive to patient preferences, needs and values, and ensures patient values guide all clinical decisions [1]. Patient centred care is multi-dimensional. For example, Mead an Bower [2] describe patient centred care as having five dimensions including a biopsychosocial perspective, the patient as a person, sharing power and responsibility, the therapeutic alliance, and the doctor as a person.

Using a patient centred care approach to deliver healthcare has been shown to reduce healthcare costs while improving patient outcomes [3, 4]. Unfortunately, not all healthcare professionals have positive attitudes towards patient centred care and attitudes vary between specialties [5]. Ensuring healthcare students have positive attitudes towards patient centred care is an important starting point to increase the number of healthcare professionals providing patient centred care. However, previous studies assessing healthcare students’ attitudes towards patient centred care have found mixed results. Some show that a large proportion of healthcare students have positive attitudes towards patient centred care, [6] while others show the opposite [7].

One possible explanation for these inconsistent findings is variation in the measurement tools used to assess attitudes towards patient centred care (e.g. Patient-Practitioner Orientation Scale [PPOS], Doctor-Patient Scale) [8, 9]. Understanding the different measurement tools used to assess healthcare students’ attitudes towards patient centred care is an important first step towards summarizing the available evidence on healthcare students’ attitudes towards patient centred care. Therefore, the primary aim of this study was to describe the measurement tools used to assess healthcare students’ attitudes towards patient centred care. Secondary aims were to quantify healthcare students’ attitudes towards patient centred care and assess the respective influence of gender, profession, and study geographical location on healthcare students’ attitudes towards patient centred care.

Methods

This systematic review has been reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols (PRISMA) [10] and preregistered on Open Science Framework [11]. The PRISMA checklist is provided in Appendix 1.

Search strategy

An electronic database search strategy was developed with a health sciences librarian and searches were conducted in MEDLINE, EMBASE, CINAHL from inception until March 1, 2021, with no language restriction. The search strategy and search terms were informed by previous reviews on patient centred care [12] and healthcare students [13]. Our search strategy combined terms relating to ‘patient centred care’, ‘attitudes’, and ‘healthcare students’ and was designed to capture studies investigating healthcare students attitudes towards patient centred care as per our preregistered protocol. The full MEDLINE search strategy is available in Appendix 2. Forward citation tracking was performed in Web of Science. All studies identified by our search strategy were retrieved and managed using Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia).

Study eligibility criteria

Studies that quantitatively assessed healthcare students’ (e.g. physical therapy, chiropractic, medicine, nursing, dentistry, etc.) attitudes towards patient centred care were included. Studies that measured mixed student and professional populations were included however, only if it was possible to extract data for students separately. Studies were not excluded based on language or type of measurement, provided it was quantitative. Qualitative studies and studies including only qualified health professionals were excluded.

Study selection

Study selection was conducted in two phases: (I) the title and abstract review phase, and (II) the full text review phase. If a paper met inclusion criteria in phase (I), the full text was retrieved and reviewed for potential inclusion. Two reviewers (GB and BC) conducted title and abstract selection and full text review independently. Any disagreements were resolved by discussion or consultation with a third reviewer (JJY).

Data extraction

Two reviewers (GB and BC) independently extracted individual study characteristics. Demographic data extracted included: author name, title, date of publication, journal, location of study, year of study completion, sample size and student characteristics (age, sex, profession). Data extraction items for study aims included: name of measurement tool and subscales, exact construct, number of items, and scoring for patient centred care measures (mean and standard deviation [SD] median interquartile range [IQR], author defined proportion of students who have positive attitudes towards patient centred care), and scoring across different sub-groups (e.g. based on age, sex, profession type). Any discrepancies were resolved by discussion between the two reviewers. Study authors were contacted when relevant data was not reported. In our protocol, we planned to extract effect measures (Odds Ratios, Risk Ratios or correlation coefficients) and measures of variability for associations between various predictor variables (e.g. age, sex, profession type) and attitudes towards patient centred care. However, no included studies reported this data.

Risk of bias/study quality assessment

The methodological quality of included studies was assessed independently by two reviewers (GB and BC) using a modified version of the Downs and Black checklist (Appendix 3). We modified the original 27-item Downs and Black checklist [14] and selected 10 items that were relevant to studies assessing attitudes towards patient centred care. Selection of items to include in the modified Downs and Black checklist was decided by consensus between study authors prior to conducting the search. The individual studies were scored from 0 to 10 based on reporting clear objectives, outcomes, characteristics of included patients, findings, estimates of the random variability, actual probability values, recruitment and sample characteristics suggesting representativeness, appropriate statistical tests, and accurate outcome measures. A detailed description of the modified Downs and Black checklist is provided in Appendix 3. Any disagreements between the two reviewers were resolved through discussion.

Data analysis

Characteristics of measurement tools used to assess attitudes towards patient centred care (e.g., name of tool, measurement construct, subscales, number of items) were qualitatively summarized. Quantitative data on attitudes towards patient centred care (mean (SD) or n (%)] was pooled when studies were considered sufficiently homogenous in terms of population and measure used to assess attitudes towards patient centred care. Meta-analysis was performed using the inverse-variance method with the Hartung-Knapp adjustment for random effects models [15]. Statistical heterogeneity was assessed using the I2 statistic [16]. The I2 statistic was interpreted as might not be important (0% to 40%), may represent moderate heterogeneity (30% to 60%), may represent substantial heterogeneity (50% to 90%), or considerable heterogeneity (75% to 100%) [17]. Analyses of factors that may influence healthcare students attitudes towards patient centred care were conducted on available variables (sex, medical students only, and United States [U.S.] medical students only) to explore whether any observed heterogeneity was due to differences in sex, profession, or geographical location across studies. Meta-analysis was conducted using R statistical software (https://www.r-project.org).

Results

The initial electronic database search identified 3948 potentially eligible studies. After removing duplicates and screening studies by title and abstract, 129 potentially eligible studies for inclusion were considered and their full text was retrieved. After full text screening, 49 studies [6,7,8,9, 18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62] met the inclusion criteria and were included in the review with 20 studies (n = 8050) [6, 7, 9, 18, 25, 26, 28,29,30,31,32, 36, 39, 43, 45, 46, 49, 52, 54, 55] providing sufficient data for meta-analysis. Two studies were conducted using the same data set as other included studies therefore, we only used data from the original report in our meta-analysis [63, 64]. The PRISMA flowchart of studies through the review is provided in Fig. 1.

Fig.1
figure 1

PRISMA flow chart of the records and study selection process

Full size image

The authors of twelve studies were contacted for additional data and were ultimately excluded due to inability to retrieve data needed to determine whether they assessed healthcare students attitudes’ towards patient centred care [65,66,67,68,69,70,71,72,73,74,75,76].

Characteristics of included trials

The 49 included studies used 16 different measurement tools to investigate healthcare students’ attitudes towards patient centred care, with sample sizes ranging from 32 to 3191 students. The majority of studies assessed U.S. healthcare students’ attitudes (40.8%) followed by United Kingdom healthcare students (8%). A comprehensive description of each study is provided in Table 1. Twenty-six studies (53%) used the PPOS measurement tool while three different modified versions of the PPOS were used in one study each. The Readiness for Interprofessional Learning Scale (RIPLS), Doctor-Patient Scale, and Interprofessional Attitudes Scale (IPAS) were each used in five, four, and two studies, respectively. The Health Beliefs Attitudes survey, Nelson-Jones and Patterson Counsellor Attitude scale, Patient-Centredness Multi-Choice Questionnaire, and Tucker-Culturally Sensitive Health Care Inventory Provider form were each used in one study. Five studies used measurement tools with no name reported. A qualitative description of all the measurement tools used in the included studies is provided in Table 2.

Table 1 Characteristics of included studies
Full size table
Table 2 Measurement tools and their subscales identified in the included studies
Full size table

Mean methodological quality assessed using a modified 10-item Downs and Black checklist was 9.04 (95% Confidence Interval [CI]: 8.73, 9.35; minimum 6; maximum 10). The most commonly omitted methodological quality indicators were a lack of appropriate reporting of probability values, subjects not being representative of the entire population from which they were recruited, and participants not being representative of the population from which they were recruited. A comprehensive breakdown of the methodological quality for each study is provided in Appendix 4.

Healthcare students’ attitudes towards patient centred care

Due to limited data, we were only able to perform a meta-analysis of studies that used the PPOS (0–6 scale) to assess healthcare students’ attitudes towards patient centred care. There were 20 studies with 26 total groups included in the meta-analysis (n = 8050). Most studies analyzed medical students (n = 18) followed by a mix of healthcare students (n = 2), nursing (n = 1), physician assistant (n = 1), dentistry (n = 1), speech therapy (n = 1), chiropractic (n = 1), and physical therapy (n = 1) students. Overall, the pooled mean score on the PPOS was 4.16 (95% CI: 3.95, 4.37; I2: 100%) (Fig. 2).

Fig. 2
figure 2

Forest plot of mean pooled PPOS score and 95% CI for healthcare students

Full size image

Factors influence on attitudes towards patient centred care

Sex, profession, and geographical location were the only factors with data available to conduct analyses to address our secondary aim of potential influence on healthcare students’ attitudes towards patient centred care. Three analyses (sex, only medical students, and only medical students in the U.S.) were conducted attempting to explain heterogeneity. Eight studies reported PPOS data stratified by sex. Among these, there were 3175 total healthcare students included (1626 men and 1549 women). The total PPOS mean score was slightly higher in women (MD 0.14, 95% CI: 0.05, 0.23; I2: 80%, n = 8 studies) (Fig. 3). PPOS mean scores were similar among subgroups of only medical students with a pooled mean score of 4.13 (95% CI: 3.85, 4.42; I2: 100%, n = 13 studies with 18 total groups) (Fig. 4a) and only U.S. medical students with a mean score of 4.49 (95% CI: 4.35, 4.64; I2: 95%, n = 5 studies with 7 total groups) (Fig. 4b). Hence, none of the analyses was able to substantially explain the heterogeneity found in the meta-analysis.

Fig. 3
figure 3

Forest plot of mean PPOS score and 95% CI difference between female and male healthcare students

Full size image
Fig. 4
figure 4

a-b Forest plots of mean PPOS score and 95% CI for medical students only

Full size image

Discussion

This is the first systematic review to summarize the measurement tools used to assess healthcare students’ attitudes towards patient centred care and quantify their attitudes. There were 16 measurement tools used to assess attitudes towards patient centred care across 49 included studies with the most common being the PPOS. Women have slightly higher attitudes towards patient centred care compared to men, and medical students, particularly those from the U.S., have slightly higher attitudes towards patient centred care than healthcare students’ overall.

Patient centred care is consistently recommended in clinical practice guidelines for a variety of conditions (e.g., musculoskeletal pain, depression, end of life care etc.) [77,78,79]. Additionally, research suggests that patient centred care is associated with higher patient satisfaction [80, 81], improved patient outcomes [3, 4, 82], and lower healthcare costs [83, 84]. Unfortunately, our findings indicate that students have low attitudes towards patient centred care overall, according to the classification by Krupat et al. [80]. Mean PPOS scores should be interpreted as high (mean score > 5.00; patient centred), medium (mean score 4.57–4.99), or low (mean score < 4.57, doctor centred). Meta-analysis of studies in our review reported a total mean score of 4.16 on the PPOS. Our findings are similar to a recent systematic review that included four studies measuring physicians’ attitudes towards patient centred care using the PPOS [12]. Those four studies reported total mean PPOS scores of 3.98, 4.08, 4.55, and 4.97 [5, 81, 85, 86]. The PPOS has demonstrated acceptable validity and adequate reliability among healthcare students [70, 87, 88]. While healthcare students are learning new information during their education and have limited time to focus on other aspects of patient care, the results of our review and Pollard et al. [12] indicate that both healthcare students and professionals have low attitudes towards patient centred care. Due to the known positive effect of patient centred care on healthcare outcomes and costs, it is important to develop and test strategies to improve healthcare students’ and professionals’ attitudes towards and implementation of patient centred care.

We found that female healthcare students have higher attitudes towards patient centred compared to males, which is similar to previous studies [70]. However, the difference between males and females was small and both groups would still be classified as having low attitudes towards patient centred care. These results imply that healthcare students require training to improve attitudes towards patient centred care and special considerations may be required for male students, but the importance of the observed difference between males and females is not clear. Analysis of only medical students found similar mean PPOS score as overall healthcare students indicating that attitudes may not differ widely between healthcare professions. However, medical students from the U.S. reported higher attitudes towards patient centred care compared to healthcare students’ overall, but again, the pooled mean score of the attitudes were still considered low towards patient centred care [6, 25, 29, 31, 36]. These results imply that there may be cultural or societal differences that may influence attitudes towards patient centred care.

Future studies assessing healthcare students’ attitudes towards patient centred care should use the PPOS to allow for comparability to previous literature or aim to validate existing tools. Many studies (41%) included in our systematic review used tools that have not demonstrated validity and reliability or have been used only once, making it difficult to interpret and compare the results of studies. Studies using a different measurement tool should look to validate and compare the psychometric properties with the PPOS.

There were only self-reported measurement tools (e.g., PPOS, RIPLS, Doctor-Patient Scale, IPAS etc.) found in our review therefore, there may be a need for objective tools used to measure patient centred care. Longitudinal studies are also needed to assess whether healthcare students’ attitudes persist into clinical practice or if attitudes evolve throughout training and with years of clinical experience. Additionally, future studies should evaluate if healthcare education can positively influence and increase healthcare students’ attitudes towards patient centred care.

Our systematic review has some limitations. We found high heterogeneity in our main meta-analyses, and this could not be explained with analyses of available factors that may influence attitudes towards patient centred care. We only included studies in English, so it is possible important data from non-English articles was missed. Additionally, our electronic database search was not conducted in all available databases, such as the Education Resources Information Center (ERIC) database or grey literature, manual searching of educational journals was not conducted, nor was pursuing the publications of relevant scholars and authors was conducted. Therefore, it is possible that relevant studies were not captured. However, our search strategy was tested independently by two research librarians, reference list screening was performed, and since all studies were related to healthcare students, it is likely they would be indexed in medical and healthcare databases. Therefore, it is unlikely that relevant literature was not included. The results of medical students and U.S. medical students only should be interpretated with caution since the majority of included studies were conducted in the U.S. therefore the results may not represent non-U.S. healthcare students.

Conclusions

We identified 16 different measurement tools that have been used to assess healthcare students’ attitudes towards patient centred care, with the most popular being the PPOS. Our results suggest that healthcare students have low attitudes towards patient centred care when measured by the PPOS. There is considerable opportunity to increase healthcare students’ attitudes toward patient centred care in order to improve patient outcomes and decrease healthcare costs. Universities have a unique opportunity to shape their curriculum to emphasize features of patient centred care. Specific classes to practice, role-play, and discuss ways to increase the dimensions of patient centred care (e.g., biopsychosocial perspective, the patient as a person, sharing power and responsibility,therapeutic alliance, and doctor as a person) may allow for increased attitudes towards patient centred care by healthcare students. This increase in attitude towards patient centred care and the dimensions that encompass it may lead to a better patient-doctor relationship that has previously led to decrease healthcare costs.