Introduction

Traditionally, patient-reported outcomes (PROs) such as health-related quality of life (HRQL) have been most commonly used as endpoints in clinical trials where aggregated PRO data is used to guide improvements in clinical care. More recently, there is increasing momentum internationally for using patient-reported outcome measures (PROMs) in routine clinical practice to support and inform the management of individual patients [1, 2]. PROMs provide unique information about the impact of disease and treatment from patients’ perspectives that can complement conventional clinical measures (e.g. blood tests, functional tests, imaging) [3] and are increasingly viewed as a key component of patient-centered care [4].

Although there is variation in how PROMs are used in clinical settings with individual patients, typically this involves a patient completing a questionnaire (or set of questionnaires) that assess health-related outcomes and results are provided to the treating healthcare professional for review [5]. In clinical practice, PROMs can be used to screen for and detect problems (e.g. symptoms of depression or anxiety), monitor changes in patient outcomes over time, and promote patient-centered care by incorporating the patient’s perspective into clinical decision-making [6].

Despite proven benefits for patient–clinician communication and patient satisfaction [7], several systematic reviews indicate mixed evidence regarding whether routine assessment of PROs in clinical practice improves patients’ health outcomes [7,8,9,10,11]. Given these variable findings, a better understanding of end-users’ experiences of using PROMs to inform patient care is needed to improve their future utility. Previous reviews have identified barriers and facilitators to using PROMs in clinical practice [12,13,14] and examined how PROMs support patient care [15]. A deeper understanding of how patients and clinicians perceive the clinical utility of PRO data based on their own experiences may provide unique insights into specific factors that promote or hinder their use in clinical decision-making. To explore this issue in-depth, we conducted a systematic review of the qualitative literature to examine end-users’ perceptions of the utility of using PROMs in diverse medical contexts. Psychiatric settings were excluded from this review as use of PROMS in these settings has been comprehensively reviewed elsewhere [16]. Our specific objective was to examine the perceived benefits and limitations of using PROMs in medical clinical practice settings from both patient and clinician perspectives, based on their real-life experience.

Methods

This review focused on qualitative studies of experiences of using PROMs in clinical practice. We chose to limit our review to qualitative studies because qualitative designs allow for more in depth exploration of patient and clinician perspectives on the benefits and limitations of using PROMs “in clinic”. From hereon, we use the term “in clinic” to refer to use of PROMs in clinical practice to support individual patient management. This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (see Supplementary File 1) and Centre for Reviews and Dissemination (CRD’s) guidance for undertaking reviews in health care [17].

Electronic searches

We searched MEDLINE, EMBASE, and PsychInfo databases from inception to May 25th 2021. Our search strategy included terms for “patient-reported outcome”, or “quality of life” or “symptom report” and “clinical setting”, or “practice”, and qualitative methods (see Supplementary File 2 for full search strategy). We also checked reference lists of included studies and relevant review articles for potentially relevant studies.

Study selection and eligibility criteria

Eligibility criteria were:

  • qualitative study design (i.e. interviews, focus groups or open-ended survey questions); mixed method studies were considered if qualitative data collection was included; and

  • focus was on using PROMs in clinic and explored patient and/or clinician perceptions of the benefits and/or limitations of PROMs in managing individual patient care.

Studies were excluded if they:

  • used quantitative methods;

  • focused on PROM development, validation, or selection for use in clinic;

  • used patient-reported experience measures for quality improvement purposes;

  • only asked about hypothetical benefits/limitations, not based on actual experience;

  • were non-English or conference abstracts; or

  • were in psychiatric settings.

All retrieved titles and abstracts were screened for eligibility by one reviewer (RC)Footnote 1 and 25% cross-checked at random by a second reviewer (AJ) [18]. If abstracts met eligibility criteria or relevance was unclear, full texts were obtained and reviewed independently by two reviewers (RC and AJ). Disagreements were resolved through team discussion.

Quality assessment

The quality of the reporting of included studies was assessed using the Consolidated Criteria for Reporting Qualitative research (COREQ) checklist [19]. The COREQ checklist includes 32 items, grouped into three domains: (i) research team and reflexivity; (ii) study design; and (iii) data analysis and findings. Each item was scored as 0 = not reported, 1 = partially reported, and 2 = fully reported, with each article receiving a total quality score out of 64, converted into a percentage (see Supplementary File 3 for scoring rules for each COREQ item). Thus, higher scores indicate higher quality reporting. Two reviewers (RC and AJ) independently assessed three articles, compared assessments, and discussed discrepancies until consensus was reached. Given minimal discrepancies, the remaining articles were assessed by one reviewer (RC or AJ) and a second reviewer (RC or AJ) cross-checked assessments against original articles.

Data extraction

A data extraction form was developed that included study aim, patient characteristics (e.g. population/disease type, sample size, age, gender), clinician characteristics (e.g. specialty, sample size, gender, age, years of experience), study design (e.g. focus groups, interviews, mixed methods), PROM used, purpose of PROM use, mode of administration (i.e. paper or electronic), and textual data regarding perceived benefits and limitations of using PROMs in clinic from patient and clinician perspectives. One author extracted data (RC or AJ) and a second author (RC or AJ) cross-checked extractions against the original article for accuracy.

Data synthesis

Thematic analysis [20] was conducted using NVivo 12. Two authors (RC and AJ; both post-doctoral researchers with expertise in PRO methodology) coded the textual data extracted from the results sections of included studies line by line to inductively identify preliminary concepts, annotating whether textual data was specific to clinician or patient perceptions. Following initial coding, both authors looked for similarities and differences between concepts and grouped them into descriptive hierarchical themes (i.e. subthemes and themes). A third author (CR) also read the articles and reviewed the descriptive themes to ensure they accurately reflected data reported in included studies. Aggregated findings across studies were grouped and summarized under each descriptive theme.

Results

Summary of included studies

Searches yielded 2388 abstracts, of which 161 were potentially relevant and 52 articles reporting on 50 studies met inclusion criteria (see Fig. 1). Included studies used focus groups (n = 15), interviews (n = 24), a combination of interviews and focus groups (n = 7), mixed methods (n = 2), or qualitative data from open-ended survey questions (n = 2) (Table 1). Studies were conducted in primary care (n = 22), hospital (n = 9), outpatient clinics (n = 9), palliative care (n = 4) or multiple settings (n = 5) and reported clinicians’ (n = 26), patients’ (n = 7) or both patient and clinician experiences (n = 17). PROMs were administered electronically (n = 26), via paper (n = 12) or either (n = 4); eight studies did not report mode of administration. Total sample size across studies was 1256 clinicians and 375 patients; four studies did not report sample size for patients, clinicians or both. Studies used PROMs in diverse patient groups, including cancer, diabetes, arthritis, HIV, asthma, Parkinson’s disease, kidney disease, among others. PROMs most commonly used (i.e. in three or more studies) included the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30), the Patient Health Questionnaire-9 (PHQ-9), and the Edmonton Symptom Assessment Scale (ESAS).

Fig. 1
figure 1

Flow chart of study inclusion

Full size image
Table 1 Characteristics of the included qualitative studies
Full size table

Methodological quality of included studies

The quality of included studies was highly variable with quality scores ranging from 8 [21] to 69% [22]. Notably, none of the COREQ items were fully reported in all studies (see Fig. 2). However, the majority of studies reported the number of participants (90%) and methodological orientation underpinning the study (60%); displayed consistency between data presented and findings (77%); and clearly presented major themes (69%). Few studies included interviewer characteristics such as gender, possible bias, and nature of the relationship with interviewee (8%); whether anyone else was present during the interview/focus groups besides the researcher (10%); and whether transcripts were returned to participants for comment (6%).

Fig. 2
figure 2

Quality of reporting across included articles (n = 52) per COREQ item

Full size image

Synthesis

Seven themes were identified: Active patient involvement and partnership; Focus of consultation; Quality of care; Standardized monitoring of patient outcomes; Patient–clinician relationship; Lack of valuable information; and Suitability for all patients. Below we describe the subthemes pertaining to each theme, which cover perceived benefits and limitations of using PROMs in clinic, or a mixture of both. Illustrative quotes are included in Table 2.

Table 2 Illustrative quotes
Full size table

Active patient involvement and partnership

Enables greater awareness and reflection

Both patients and clinicians reported completing PROMs helped patients identify their needs and symptoms [23,24,25,26,27,28,29]. Some patients reported PROM completion increased their self-awareness by “forcing them to think” and identify specific problem areas [22, 30,31,32,33,34], whereas clinicians reported PRO data helped reflect on patients’ progress and improved their understanding and awareness of patients’ physical and mental health [35,36,37,38].

Encourages patient involvement

Both patients and clinicians reported PROM completion improved communication by helping patients recall symptoms to discuss with their clinician [27, 31, 32, 39] and clinicians to pinpoint issues to discuss with their patients [40, 41]. This enhanced communication was reported by both as encouraging patients’ to work collaboratively with clinicians and be more involved in their care [24, 29, 42, 43].

Facilitates goal setting

PRO data was reported to facilitate short- and long-term goal setting by patients and clinicians [24, 44]. Clinicians found patient-reported functional data particularly useful for setting functional goals throughout treatment [21]. Both patients and clinicians reported PRO data helped to motivate and “reinforce positive change when patients’ symptoms were on track” [34, 35]. Some clinicians stated individual items (rather than domain scores) were especially useful for setting goals with patients [28].

Influences honesty

Patients and clinicians had mixed opinions about patients’ honesty when completing PROMs. Some patients reported the impersonal nature of PROMs promoted honesty whereas others reported it facilitated dishonesty [25, 27, 29, 45]. The degree of honesty may be dependent on the type of PROM, with some patients expressing they were more dishonest when completing PROMs assessing depression, to avoid unwanted treatment or being judged [29]. Clinicians similarly expressed mixed views, with some reporting patients were more honest because they did not need to please clinicians directly while others reported patients hid symptoms and responded in socially desirable ways [22, 40, 45,46,47].

Permits discussion of sensitive topics

PROMs were reported to help patients and clinicians discuss difficult or embarrassing topics during the clinical encounter (e.g. sexual or mental health issues) [24, 38, 46]. Some clinicians reported completing PROMs gave patients “permission to talk” about sensitive issues because PROMs covered questions not usually asked by clinicians [48]. Others stated using PROMs helped patients with communication difficulties to express concerns in numbers rather than words [22, 49].

Focus of consultation

Helpful as a screening tool

Receiving PRO data before the consultation helped some clinicians better prepare by providing an overall impression of their patient’s condition and identifying issues to focus on during the consultation [22, 50]. This reduced time needed to take detailed histories, allowing them to focus immediately on patient priorities [42].

Prioritizes patients’ needs

Many patients and clinicians reported using PROMs helped prioritize patients’ needs and identify problems that may have otherwise been overlooked [22, 29, 33, 37, 40, 42, 44, 46, 48, 51,52,53]. It also helped clinicians identify areas in which patients were doing well, enabling them to focus on other issues important to the patient [52]. Some clinicians stated PRO data did not change the questions they asked, but rather the order in which they asked questions [22, 52], preventing important issues from being left till last [28, 33, 52]

PRO data was reported to trigger conversations about patients’ priorities for care and treatment [31]. This was appreciated by patients, because patients and clinicians can differ in what they perceive as most important in relation to care [31, 42]. One clinician described patients as looking “relieved we’re talking about what is most important to them” [28]. Some clinicians reported PRO data captured information essential for shared decision-making and helped clinicians understand patients’ treatment preferences [42].

Conversely, some clinicians reported using PROMs negatively shifted the focus of consultations away from the patient’s main health complaint, towards other less important symptoms, particularly when PROMs were symptom-specific (e.g. focused on pain) [44]. Those clinicians felt this resulted in them having to manage lower priority problems and reduced the time to deal with other important issues [44, 54].

Provides one piece of the picture

PROMs were perceived by some patients and clinicians to reduce complex conditions to numeric scores and only provide one piece of the picture [29, 54]. Several clinicians emphasised the importance of using PRO data in combination with clinical skills and questioning patients further about their broader life context [55].

Structures consultations and improves efficiency

PRO data helped several clinicians structure consultations by highlighting areas to focus on Refs. [27, 37, 46, 48]. PROMs administered electronically were reported especially efficient because they provided information at a glance, enabling clinicians to prepare for consultations time-effectively [25, 27, 31, 46]. The process of completing PRO assessments also helped patients feel better prepared, resulting in a more efficient consultation [32].

Quality of care

Assists diagnosis and enables tailored care

PRO data assisted some clinicians to make accurate diagnoses by identifying exactly where problems lay [29, 54]. Others reported it helped determine goals of treatment, resulting in more individualised care plans [21, 25, 51, 53, 56]. Patients similarly believed that reporting symptoms systematically assisted their healthcare team to make decisions about their care [52].

Ensures holistic care

Several clinicians stated PROMs covered issues essential for holistic care such as patients’ quality of life [36, 42]. By providing information about different aspects of quality of life (e.g. emotional, social and spiritual), PRO data helped clinicians understand patients’ lives more holistically [26, 38, 48, 57, 58].

Can inaccurately estimate the problem

Some clinicians reported PRO data inaccurately estimated patients’ problems by over- or underestimating symptoms [27, 29, 40, 44, 55]. Some were concerned PRO data led to “false positives” and encouraged patients to “find” a complaint, pushing clinicians to intervene even if treatment was not a priority [44]. Others were concerned PRO data underestimated patients’ problems and reported cases of patients scoring low on symptoms (e.g. dyspnea) even though they exhibited signs of more severe symptom burden [49, 59, 60]. Several patients also expressed difficulty quantifying the level of their symptoms [25].

Prompts appropriate action

Many patients expected PRO data to prompt clinicians to take action to manage their symptoms [27, 32, 39]. When PRO scores did not prompt any action, patients wanted reassurance their scores had been considered when developing treatment plans [26, 31, 39]. Conversely, other patients were not aware that their PRO data could inform treatment decisions [39].

Many clinicians reported PRO data prompted them to address symptoms [29, 31, 46, 55], while others expressed concern that patients may expect PRO scores to prompt action. Some found it difficult to act on PRO data and worried they were expected to address all the issues reported, even though they were not equipped with adequate resources to manage them [38, 40, 41, 45, 46, 49, 50]. Others reported the demands induced by PROMs was a stress factor that had a negative bearing on their job satisfaction [37, 38].

Standardized monitoring of patient outcomes over time

Helps determine effectiveness and side-effects of treatment

Routinely assessing PROs in clinic helped some clinicians assess whether treatment was achieving desired outcomes, resulting in modifications to treatment plans if symptoms were not improving [33, 35, 42, 51, 52, 61]. Others reported it helped monitor acute and late effects of treatment [52, 61] and determine whether supportive care was required [62].

Useful for monitoring changes and tracking progress

Regular PRO assessment helped clinicians monitor changes and track progress in patient outcomes over time [49, 61]. Some clinicians reported showing patients trends in their PRO scores over time helped illustrate improvements in their condition [24, 26]. Patients reported seeing their PRO data helped to “see how far they’ve come and how far they needed to go” [63]. While some patients found it encouraging to see symptoms improve over time, others found it distressing when symptoms worsened [23].

Patient–clinician relationship

Provides reassurance that clinicians care

Some patients reported completing PROMs helped improve the patient–clinician relationship by reassuring patients that clinicians cared [32, 40]. It helped some patients feel they were being taken seriously and induced a feeling of not being left alone [29]. Several clinicians reported using PROMs improved the patient–clinician relationship by demonstrating clinicians’ interest in all aspects of patients’ well-being [26, 35, 36, 42, 46, 53].

Inhibits interaction and rapport

Others expressed the opposing view, namely that PRO assessment inhibits interaction and rapport. Some patients found PROM assessment impersonal because it gave the impression clinicians were too busy to talk to them [26]. As a result, one patient stated they wanted to “go back to the days when doctors were concerned about human beings and not numbers” [26]. Others worried PROM assessment would reduce the time available to talk with the clinician [35, 52].

Several clinicians were similarly concerned PROM assessment would be perceived as impersonal and harm the patient–clinician relationship [37, 40, 47, 49, 64]. One clinician argued that “patient-centered care should rely on conversation and focus on knowing the human, not data” [50]. Others feared using PROMs would lead to a de-skilling of clinicians and produce a “generation of doctors driven by ticking boxes” [29].

Lack of valuable information

PRO data is not specific enough to be clinically meaningful

Generic PROMs (e.g. the SF-36) were reported by some clinicians to be less clinically meaningful than disease-specific PROMs because they provided information on outcomes outside clinicians’ control [26]. Some clinicians were also critical of how PRO data was reported, stating that scores of individual items were more clinically meaningful than aggregated multi-item scores because items enabled identification of specific issues of concern [52].

Both clinicians and patients reported patients had difficulty answering questions in some PROMs because questions were unclear or irrelevant [25, 39, 46, 57, 59]. Some patients also struggled to choose a response because options provided were not specific enough [25, 39].

Provides redundant information

Some clinicians stated PROMs provided redundant information because they covered questions already asked by clinicians and added no new information [52]. Perceived usefulness of PROMs was somewhat dependent on clinicians’ years of experience; PROMs were considered most useful for less experienced or less confident clinicians [26, 29, 47, 51, 54].

Suitability for all patients

Several clinicians and patients believed PROM assessment was not universally suitable for all patients [26, 31, 44]. Some clinicians reported very sick or highly distressed patients were unable to concentrate on completing PROMs [45, 55, 58]. Others felt patients in palliative care were so used to their symptoms they had difficulty quantifying them, or so heavily sedated they were unable to reliably respond to questions in PROMs [56]. Several clinicians reported PROM completion was too complex for elderly, cognitively impaired or low literacy patients [26, 33, 53, 65].

Discussion

This review identified 52 articles reporting on patient and clinician experiences of using PROMs in clinical practice to inform the management of individual patients. Synthesized evidence indicated both patients and clinicians reported many benefits of using PROMs in clinic. These include five key benefits: (1) promoting active patient involvement in their care by facilitating goal setting, and permitting discussion of sensitive topics; (2) enhancing the focus of consultations by prioritizing care around patient needs; (3) improving quality of care by enabling tailored, holistic care and prompting appropriate action; (4) enabling standardized monitoring of outcomes over time to monitor PRO changes and track progress; and (5) enhancing the patient–clinician relationship by reassuring patients that clinicians care. A number of limitations were also identified such as the capacity for PROMs to negatively shift the focus of consultations and reduce quality of care by inaccurately estimating symptoms and raising expectations for care that exceed clinicians’ resources. In some studies, PROMs were reported to inhibit the patient–clinician relationship, lack clinically meaningful information and were not considered suitable for all patients.

Although patients and clinicians tended to converge on the perceived benefits of using PROMs, several limitations were uniquely reported by clinicians (e.g. PROMs provide redundant information and negatively impact on the focus of consultations). Given that clinicians play a key role in the utilization of PRO data, these negative perceptions may hinder the potential for PRO data to drive clinical decision-making and improve patient outcomes. These findings suggest that shifting clinician attitudes by providing training and education on the added value of PRO data, may help to improve the effectiveness of using PROMs in clinic. They further highlight the importance of emphasising that PROMs administration is intended to promote high quality standardized patient-centered care and enhance communication with patients’ about their needs and concerns rather than replace patient–clinician conversations.

The themes identified were remarkably consistent across diverse health conditions and clinical settings. The most common themes identified across ten or more health conditions were enabling greater awareness and reflection, useful for monitoring changes and tracking progress, influencing honesty in disclosure and inhibiting interaction and rapport. The least common themes, emerging across only five or fewer health conditions, were helpful as a screening tool, structures the consultation and improves efficiency and provides redundant information. The consistency of themes across clinical settings was also noteworthy, with all themes emerging across three or more different clinical contexts. The biggest discrepancies were observed within palliative care settings, where several themes did not emerge (i.e. facilitates short & long term goal setting, helps determine effectiveness of treatment, helpful as a screening tool, PRO data not specific enough to be clinically meaningful, and provides redundant information).

Importantly, the identified limitations highlight why use of PROMs in clinical settings may not always improve patient outcomes and indicate important considerations to be addressed in the design and implementation of future PROM initiatives. In line with the International Society for Quality of Life Research (ISOQOL) guidance [66], these findings emphasize the importance of choosing appropriate PROMs for the patient group and clinical context, reporting PRO results to clinicians in an easily interpretable and clinically meaningful format, and developing feasible strategies or guidance for responding to issues identified by PROMs. It is essential that PROMs support clinicians to provide enhanced care, rather than add to job demands. While patient engagement with PROMs is critical, clinicians are responsible for using PRO data to inform patient care and require adequate support and training to realize their full potential. Co-designing PROM initiatives with patients and clinicians may help ensure clinical relevance and feasibility.

A key strength of this review is that it synthesizes qualitative evidence on the clinical utility of PROMs among a range of patient groups, across diverse clinical settings and countries, from both patient and clinician perspectives. Although other qualitative reviews have also examined end-users' experiences of using PROMs in clinical practice, they focused mainly on barriers and facilitators to PROM implementation either from clinicians' perspectives [12] or within a particular clinical setting [13]. The themes identified in our review closely align with key findings from a realist synthesis of qualitative and quantitative data [15], which also found PROMs facilitated reflection and gave patients permission to raise issues with clinicians but could sometimes constrain rather than support communication [15]. This review supports and extends these findings by also identifying several additional themes common across diverse health conditions and clinical settings such as facilitating short and long term goal setting and influencing honesty during consultations. This review also provides some insight into findings from quantitative systematic reviews [7,8,9, 11, 12] which found mixed evidence for the effectiveness of using PROMs in clinic by identifying several limitations such as the capacity of PROMs to underestimate problems or lack clinically meaningful information, which may have contributed to variability in patient outcomes across studies.

This review also has some limitations. Our search strategy did not include terms for PROMs often used in the field of psychiatry and clinical psychology, such as routine outcome monitoring (ROM), clinical feedback (CF), measurement feedback system, and feedback-informed treatment. As such our findings may not be generalizable to these settings. A systematic review of qualitative studies examining patient experiences of ROM/CF systems as part of psychological therapies within mental health settings found some benefits similar to those identified in this review, such as empowering patients to be more involved in their care, encouraging a collaborative practice by helping patients to set goals, track progress, reflect and become more self-aware [16]. However, that review also identified some negative perceptions that did not emerge in our review, such as suspicion towards service providers including concerns about confidentiality and fears that patients’ PRO data could be used against them to limit or deny access to services [16]. Privacy concerns were reported in two studies included in our review [26, 31], with some patients expressing fear about whether insurance companies would have access to their data. However, as this concern only emerged in two studies it did not warrant inclusion as a major theme or subtheme in our review.

We acknowledge that our search strategy may also have missed other relevant papers. This arises due to different terminology heritages and conventions across disciplines, creating a complex problem for cross-disciplinary systematic reviews. Arguably this problem could be solved by identifying an exhaustive multidisciplinary set of terms for PROMs, but its use would likely retrieve thousands of abstracts to screen, a task which may not be feasible for many research teams due to resource constraints. Another limitation is that the methodological quality of included studies was highly variable. Several studies lacked detail about their methods which may affect the credibility of some findings. In addition, it was not possible to synthesise data by sample characteristics, either because patient and clinician characteristics were poorly reported or findings were not reported by sample characteristics in the original studies. This is a missed opportunity as it prevents gaining insight into specific patient and/or clinician characteristics that promote or hinder the use of PRO data in clinical decision-making. We also did not assess inter-rater consistency during screening and full text review but any discrepancies were discussed until consensus was reached. Finally, this review was also limited to full text articles published in English and may have excluded informative studies published in other languages and formats.

This review identified some critical gaps in evidence. Only one study reported on the experiences of ethnic minorities [57] and few studies included patients with low socio-economic status or low literacy. As a result, the potential benefits and limitations of using PROMs in clinic is not well known from the perspective of these vulnerable patient groups. Further qualitative research is needed to gain insights into the unique experiences of these and other more socially disadvantaged patient groups to ensure future PROM initiatives are designed to be equitable and inclusive.

In sum, the present review identified many benefits but also highlighted several limitations of using PROMs in clinical practice from both patient and clinician perspectives. Although patients and clinicians agreed on many of the perceived benefits, several of the limitations such as concerns about the validity of existing PROMs, clinical relevance or added burden on job demands were only reported by clinicians. In order for PRO data to be useful and improve patient outcomes, it is essential that these limitations are addressed in the design and implementation of future PROM initiatives.