This study explored which quality indicators patients find most important when choosing a hospital for THA/TKA and whether patient subgroups differed in which indicators they find important. By means of ranking indicators, patients indicated the importance of different patient experience indicators (CQI themes derived from the Dutch Consumer Quality Index questionnaire), clinical performance indicators, and indicators about hospital services. We demonstrated that patients perceived the conduct of doctors, the existence of procedures to prevent adverse effects of thrombosis and information about the specialist area of orthopaedists as the most important patient experience indicator, clinical performance indicator and indicator about hospital services, respectively. When presenting comparative healthcare information, it seems that these quality indicators are essential to incorporate. Our results also showed that patients of different subgroups and patients within subgroups sometimes differed in the importance attached to certain quality indicators. Although implementation can be difficult, website builders could consider supplementary options to present tailored comparative healthcare information. Hopefully, this will stimulate the use of comparative healthcare information by patients, which is limited up to now [
3].
Due to the explorative nature of this study, definite conclusions as to which indicators to present cannot be drawn. More extended research based on the current results is needed to select the indicators to be presented for THA/TKA patients. Subsequent studies should preferably use experimental methods and include a larger group of patients to further investigate the perceived importance of the various quality indicators.
Discussion of the results
Although in this study no direct comparison was made between the three different types of quality indicators, it seems that both interpersonal (i.e. conduct of doctors) and more technical aspects (i.e. prevention of adverse effects of thrombosis, specialist area of orthopaedists) are important for patients, which is in line with research of Wensing et al. (1998) [
20]. Previous research has found mixed results concerning healthcare preferences: in some studies it was found that patients value the technical aspects more than interpersonal skills and service aspects [
7,
23], whereas other studies showed the opposite pattern [
25,
36]. The type of research, the patient group involved and the kind of healthcare choice may all affect patients' priorities. Evidence that preferences change over time [
37] and the theory of constructed preferences [
38], addressing that patient's preferences are often constructed during the decision making process, are reasonable explanations for the differences in findings between studies. That patients' priorities differ from person to person and can change from time to time, shows the complicated process to provide relevant comparative healthcare information that suits different patients.
The fact that patients can value different types of information as important (e.g. interpersonal and technical aspects), can increase the cognitive burden to make a deliberate hospital choice [
16,
28]. Patients have to weigh different indicators and have to make trade-offs in their choice. The decision making process will always require difficult cognitive processes, especially when contradictory information is involved. It is known that tailoring information can ease the cognitive burden as less information processing is required, because only the relevant aspects have to be taken into account in the decision making process [
16].
Distance to the hospital, reputation of the hospital, waiting times, hospital facilities (e.g. revalidation options, extended visiting hours) and experiences of others were mentioned as other important indicators when choosing a hospital, which is in line with previous studies [
22,
39‐
42]. The integration of this kind of information into comparative healthcare information could thus be considered. However, the amount of information on web pages needs to remain manageable for patients. We would therefore advocate further research about the value of these indicators for patients' choice.
Our results show that patients' characteristics may to some extent determine which quality indicators patients find important, which corresponds to findings from previous research [
21,
42‐
44]. Gender and educational level of THA/TKA patients influenced the importance attached to the indicators information about new medication and the number of yearly performed total knee- or hip replacements among adults. In addition, age, gender, and educational level affected the value attached to information provision before surgery and the presence of procedures to prevent adverse effects of thrombosis. Although no significant differences were found for this latter result, - perhaps as a result of our small sample size -, these results indicate that different patients can have different preferences, which would challenge the 'one size fits all' approach of comparative healthcare information.
An interesting finding is that higher educated patients ranked the yearly performed total hip- or knee replacements among adults as more important than lower educated patients. The number of performed surgeries often is associated with the outcome of the surgery, by influencing the complication or mortality rates [
45‐
47]. After all, the more experience doctors have, the less chance medical mistakes occur. The question rises whether people with a lower educational level interpret this indicator as such. More qualitative research is needed to provide insight into how people interpret this kind of quality indicators.
Given these results, an interesting thought is to what extent the perceived relevance of indicators by (future) patients should drive the selection of quality indicators for comparative health care information. One could argue that crucial information, like the number of performed surgeries, should be available for every patient even if patients do not perceive it as one of the most important indicators. When such crucial information is only provided to patients who attach value to it, this may possibly result in inequities in the healthcare system. Ideally, comparative healthcare information consists of a mixture between indicators perceived as important by patients, and crucial indicators known to be related to treatment outcomes perceived as important by experts.
A few significant differences in between-participants variance scores were found between subgroups, indicating that individuals
within some patient subgroups also were less unanimous about the importance of certain quality indicators. It was notable that patients with a good to excellent health and a low educational level were less unanimous about the importance of conduct of doctors, which overall was ranked as the most important patient experience indicator. That individuals in some patient subgroups differed about the importance of indicators, stresses the relevance of a tailor-made presentation approach for comparative healthcare information [
16,
48], but also shows the difficulty to create this kind of information.
Strengths & Limitations
An important feature of this study is that participants ranked different quality indicators instead of rated the importance of these indicators. By ranking indicators, patients must explicitly weigh what they find more and less important. By rating indicators, in contrast, patients can rate every indicator as equally important. So, by using the ranking method, the results provide more insight into what patients find most important. Another strength of this study is that we asked participants which quality indicators they find important when choosing a hospital, rather than we focused on which quality indicators they find important in general. As a result, the findings are more likely to improve comparative healthcare information in terms of relevance for patients' hospital choice. A final strength is that we included real patients who face or have faced a real hospital choice.
Although this explorative study adds to our understanding of what patients find important when choosing a hospital for THA/TKA, it also has several important limitations. First, the small sample size decreased the statistical power of our study and, combined with the multiple tests performed, this may lead to potential sources of both systematic and random errors. By using a P of 0.01, however, the chance that results are capitalizing on chance only is smaller.
Second, there are some limitations related to the representativeness of the sample. The majority of the participants had enrolled themselves in this study. Given this self-selection, participants of this study may be more interested in the topic of comparative healthcare information than the average THA/TKA population. When background characteristics of our participants were compared to participants of a nation-wide study using the CQI Hip Knee Questionnaire, it appeared that our participants were younger and had a higher educational level than the participants of the CQI Hip Knee-survey. The majority of participants also consisted of patients who had already undergone surgery. Patients who still have to undergo surgery may evaluate other quality indicators as important, because the phase or severity of the disease can determine patients' preferences [
7]. In sum, it is unclear to what extent the results can be generalized to a larger group of THA/TKA patients.
Finally, besides the usual limitations inherent to an explorative study (e.g. potential for spurious relationships), there are some other methodological issues. The use of a ranking method forces people to make choices and this creates a hypothetical situation, since people can decide in real life to make no choice at all. In addition, by using separate ranking assignments for the three types of quality indicators, it was not possible to compare across the three dimensions of the quality indicators. The use of different ranking assignments, however, was a deliberate choice to decrease the cognitive burden for participants to fulfil the assignments. At last, this study was limited to three types of quality indicators. Other indicators also can be important for patients when choosing a hospital, as was illustrated by the answers of participants to our open ended question. Taking all these limitations into account, caution is warranted by the interpretation and generalization of the results.
Implications
Despite the explorative nature of our study, our findings are already informative and of importance for builders of websites who develop comparative healthcare information for patients in general and, more specifically, for THA/TKA patients. Providers of comparative healthcare information often select the indicators that will be presented themselves, without taking the needs and wishes of patients systematically into account. However, involving patients, the users of the information, in the process of designing information is essential in making comparative healthcare information more relevant.
To some extent, our recommendations seem contradictory: on the one hand we advocate to only present concise information, on the other hand we suggest the presentation of tailored information. However, both aims can simultaneously be accomplished. This study gives a first impression of which quality indicators are most important from the patient's perspective: the conduct of doctors, the existence of procedures to prevent adverse effects of thrombosis, and the specialist area of orthopaedists. Making use of a succinct overview on the first web page for this kind of information will prevent that THA/TKA patients feel overwhelmed by the information provided. When considering supplementary tailored information about THA/TKA, the use of deep-linking and selection tools can be considered [
5,
28,
49]. An option to consider is that patients themselves select the quality indicators they find important when choosing a hospital. The number of selected indicators can differ from person to person, adjusted to someone's own wishes. In this way, patients have an active role in their own information supply. The challenge ahead is to create a balance between the provision of concise, crucial information on the one hand, and to create supplementary tailored information that suits the target group on the other hand.
For future research we would recommend to examine which quality indicators are important for other patient groups when choosing a healthcare provider, in order to develop relevant comparative healthcare information for them as well. In our opinion, using different methods to elicit patient preferences is preferable. The use of the ranking method to explore patients' preferences is a good starting point, complemented with other research (e.g. discrete choice experiment or qualitative research) to profoundly explore patients' preferences. Although the central focus of this study concerned which aspects have to be presented for consumers, we acknowledge that the presentation approach is just as important [
8,
16] for making comparative healthcare information successful.