Home > Using Research > What types of research can provide evidence about patient views?

Published and unpublished research about patient views can come from a single study or from a systematic review, and can include of a variety of study designs for different purposes:

  • qualitative research, such as interviews and focus groups
  • surveys
  • comparative studies (non-randomised and randomised)
  • studies providing utility and non-utility estimates for an outcome, and
  • studies that determine minimally important differences (MIDs) in an outcome.

Qualitative research studies

Information about patient views will often come from qualitative research studies, such as interviews and focus groups. This evidence can highlight areas of concern to patients, which may inform the scope of the guideline, the significance of guideline questions, the relative importance of outcomes, and acceptability of interventions. In turn, these concerns can be considered by the guideline group when deciding which questions and outcomes to address, and when incorporating values and other factors into recommendations. For example, in a guideline about different care models for people with haemophilia, a qualitative study, consisting of interviews with patients and the results of a thematic analysis, was published (Lane et al. 2016). The study reported important aspects to patients related to different models of care, 2 of which were equal access to comprehensive models of care, and the perception that integrated care is better. Consequently, when making the recommendation for the integrated care model, the guideline panel included additional guidance about overcoming system level and patient level barriers to ensure equal access.

Surveys

Surveys can provide valuable information about patient views. In particular, surveys are often used to ascertain the important questions patients have about a topic. An example of a primary study is an online self-administered survey of members of the Canadian Osteoporosis Patient Network, who were asked what priority interventions should be covered in a new guideline for osteoporosis management (Morin et al. 2020). Over 1,000 people rated interventions, such as physical activity or nutrition, from ‘1 = not important’ to ‘5 = critical’. The ratings were used to prioritise topics for the new guideline. Published surveys can also be used to determine the important outcomes to review for a particular question. The Core Outcome Measures in Effectiveness Trials (COMET) initiative is a database of studies, including surveys, that identifies outcomes to measure and report in trials for different health conditions. The surveys can be used to inform guideline developers about which outcomes should be covered in the systematic reviews and should be weighed when making the recommendations.

Comparative studies

Although quantitative research is typically used to determine the effects of interventions and tests, comparative studies can also include outcomes directly related to patient views. Acceptability of an intervention can be assessed in participants and compared between a group that receives the intervention and a group that does not. Whether patients experience barriers or other challenges because of costs, resources, equity or feasibility issues can also be measured and compared between groups. Both randomised controlled trials and comparative non-randomised studies may provide this evidence. For example, for a World Health Organization (WHO) guideline with recommendations on treatments for precancerous lesions, the guideline group was considering whether to recommend 1 surgical treatment (loop electrosurgical excision procedure) rather than another (cryotherapy). A randomised controlled trial measured important health outcomes with either treatment and also how many participants felt the procedure they received was acceptable (Chirenje et al. 2001). Acceptability was found to be similar and contributed to the decision of the guideline group to suggest treatment with either procedure.

In another WHO guideline, recommendations for treatment of chlamydia were made. Randomised controlled trials provided information about the difference in effects of various treatments and adherence to those treatments. The information about adherence was used to inform decisions about patient preference for single-dose compared with multi-dose regimens (Hillis et al. 1998)

Studies providing utility and non-utility estimates

Studies may also quantify the value placed on an outcome as utility and non-utility estimates. The use of these estimates in guideline development is relatively new and methods for incorporating this evidence are still being developed (Zhang et al. 2017) Essentially, studies will use different methods to measure utilities (such as Standard gamble or Time trade off) and report the utility of a health state (for example, a health outcome) on a scale from 0 (death) to 1 (perfect health). When comparing utilities for different outcomes, guideline groups could consider outcomes with lower utilities as an indicator of less desirable outcomes that may carry greater weight when balancing effects, and a wide range in a utility score as an indicator that patients may not value the outcome (or health state) similarly.

Studies that determine MIDs

Another type of study related to patient experiences and views is a study that measures MIDs, defined as the smallest change, either positive or negative, that patients perceive as important (Schünemann and Guyatt 2005). This information can be used by guideline developers to interpret the magnitude of the effect of an intervention on an outcome. For example, in a guideline comparing a surgical procedure with non-operative treatment for shoulder pain, studies were available that determined the MIDs for instruments used to assess shoulder pain, function, and health-related quality of life (Hao et al. 2019). When the guideline panel had to decide how large the benefits of the surgery were and how large the harms were, they used the MIDs identified by patients. They determined that the magnitude of the benefits of surgery were less than the MIDs, and magnitude of the harms were greater than the MIDs, and therefore recommended against the surgery.

Systematic reviews

Finally, guideline groups can use the evidence from any of the above studies individually or synthesise such studies. If a systematic review of these studies is already available, that may be preferable because it reduces time and resources necessary to gather evidence about patient views. It could also save resources if there is a diverse or large body of evidence already available. When searching for systematic reviews, guideline groups should be aware that there is no standard for reviews of patient views, and groups will likely need to delve into the reviews for the details. Systematic reviews will have different purposes and therefore specific inclusion and exclusion criteria that guideline developers will need to assess as relevant before using the reviews. Systematic reviews may

  • cover broad questions about patient views related to priority questions (see the series about how to use this type of review [Downe et al. 2019])
  • cover all factors related to patient views, such as values, and acceptability and equity issues (see the series on how to use this type of review [Lewin et al. 2019])
  • cover implementation issues (see the series about how to use a review for this topic [Glenton et al. 2019], and box 1 for an example), or
  • restrict types of study designs included, such as qualitative research (see the series on how to use this type of review [Lewin et al. 2019]).

BOX 1: A systematic review of patient values and preferences

The American Society of Hematology developed guidelines for management of venous thromboembolism (VTE) disease. There are important trade-offs in VTE management, in particular, because interventions that reduce the risk of thrombosis increase the risk of bleeding. A systematic review of patient values and preferences related to VTE was conducted (Etxeandia-Ikobaltzeta et al. 2020). Multiple databases were searched for both quantitative and qualitative studies.

When summarising the data, the authors combined the results of quantitative and qualitative studies, and also conducted a separate analysis of the themes and quotes from the qualitative studies. The evidence from quantitative studies included utility estimates for outcomes (for example, deep vein thrombosis, gastrointestinal tract bleeding), and non-utility estimates about outcome priorities (for example, VTE risk reduction), willingness to accept a particular intervention (for example, treatment burden of vitamin K agonists), treatment method preference (for example, injection compared with oral medication), and testing method preference (for example, thrombophilia testing). The qualitative studies provided important information from patients related to disease treatment benefits and burden, healthcare provider communication and relationships, awareness and perceptions of risk, and day-to-day routines. Overall, the evidence suggested that patients put higher value on VTE risk reduction than on the potential harms of the treatment, and likely prefer oral medication rather than subcutaneous medication.

The guideline panel used this information to inform the:

• values placed on outcomes and whether the values are consistent across populations, and
• acceptability and feasibility of the interventions when making the recommendations and writing additional guidance about implementation.

When systematic reviews are not available, a guideline group may decide to conduct their own systematic review. In this chapter, we do not provide details of how to conduct a systematic review, but we will describe some of the unique elements pertaining to syntheses of research about patient views in the next sections. For details about how to conduct systematic reviews that include a variety of different study designs, the resources in table 2 may be helpful.

Table 2 Resources for how to conduct systematic reviews

Type of systematic reviewGuidance for conducting the review
Review of randomised and non-randomised studiesCochrane Handbook
Synthesis of qualitative researchCochrane Handbook: Chapter 21: Qualitative evidence Additional guidance: Cochrane Qualitative & Implementation Methods Group
Synthesis specific to quantitative patient valuesGeneral guidance: Zhang Y, Coello PA, Brożek J et al. (2017) Using patient values and preferences to inform the importance of health outcomes in practice guideline development following the GRADE approach. Health Quality Life Outcomes, 15: 52
Overviews of reviewsCochrane handbook: Chapter 5: Collecting data
Rapid reviewsCochrane Rapid Reviews Methods Group