Respectfulness is one measurable and core element of healthcare responsiveness. The operational definition of respectfulness is "the extent to which health professionals and support staff meet users' expectations about interpersonal treatment, demonstrate respect for the dignity of patients and provide adequate privacy."

Objective: To examine how well respectfulness is captured in validated instruments that evaluate primary healthcare from the patient's perspective, whether or not their developers had envisaged these as representing respectfulness.

Method: 645 adults with at least one healthcare contact with their own regular doctor or clinic in the previous 12 months responded to six instruments, two subscales that mapped to respectfulness: the Interpersonal Processes of Care, version II (IPC-II, two subscales) and the Primary Care Assessment Survey (PCAS). Additionally, there were individual respectfulness items in subscales measuring other attributes in the Components of Primary Care Index (CPCI) and the first version of the EUROPEP (EUROPEP-I). Scores were normalized for descriptive comparison. Exploratory and confirmatory (structural equation modelling) factor analyses examined fit to operational definition.

Results: Respectfulness scales correlate highly with one another and with interpersonal communication. All items load adequately on a single factor, presumed to be respectfulness, but the best model has three underlying factors corresponding to (1) physician's interpersonal treatment (eigenvalue=13.99), (2) interpersonal treatment by office staff (eigenvalue=2.13) and (3) respect for the dignity of the person (eigenvalue=1.16). Most items capture physician's interpersonal treatment (IPC-II Compassionate, Respectful Interpersonal Style, IPC-II Hurried Communication and PCAS Interpersonal Treatment). The IPC-II Interpersonal Style (Disrespectful Office Staff) captures treatment by staff, but only three items capture dignity.

Conclusion: Various items or subscales seem to measure respectfulness among currently available validated instruments. However, many of these items related to other constructs, such as interpersonal communication. Further studies should aim at developing more refined measures – especially for privacy and dignity – and assess the relevance of the broader concept of responsiveness.

In 2000 the World Health Report (WHO 2000) rated the health systems of member countries on the basis of fairness, effectiveness and responsiveness. In this important document, responsiveness was defined as "the extent to which non-medical expectations of patients are met." This report highlighted the importance of meeting patients' expectations, effectively and equitably, regarding what care they should receive and when and how care should be provided. More specific to primary healthcare (PHC), a recent Canadian study identified responsiveness as one of the six domains important to the public (Wong et al. 2008).

Background

Conceptualizing Respectfulness

In 2004, we conducted a consensus consultation of 20 PHC experts across Canada on attributes of primary healthcare that should be evaluated in health reforms (Haggerty et al. 2007). Reflecting the WHO report, responsiveness was identified as one of the attributes to be measured. However, when mapped to validated instruments, the WHO definition of responsiveness proved to be very broad and overlapped with many other attributes, such as interpersonal communication and relational continuity. In addition, conceptualizing responsiveness as the generic ability to respond to a broad range of needs expressed by the patient did not seem specific enough for evaluative purposes. Consequently, the PHC experts narrowed the concept of responsiveness and pinpointed "respectfulness" as fitting more specifically with how providers respond to patients' expectations, leaving aside aspects related to the range of services available and timeliness in responding to patients' needs. In purely linguistic terms, responsiveness is related to the quality of reacting or responding quickly, whereas respectfulness relates to the quality of being courteous, humble, reverent and deferential (Canadian Oxford Dictionary 1998).

The experts agreed to define respectfulness as "the extent to which health professionals and support staff meet users' expectations about interpersonal treatment, demonstrate respect for the dignity of patients and provide adequate privacy." They unanimously agreed that this attribute is most validly evaluated from the patient's perspective.

Currently, there is little comparative information to guide evaluators in their selection of the appropriate tool for evaluating respectfulness of care. Various instruments that evaluate primary healthcare from the user's perspective address aspects related to respectfulness, although respectfulness as a construct is addressed only by one instrument, to our knowledge.

In this paper, we present and discuss results from a concurrent validation process of existing instruments that assess primary healthcare from the patient's perspective, with regard to the evaluation of respectfulness. More specifically, our objectives were to contribute to the understanding of the concept of respectfulness, given the lack of instruments available to measure it, and to explore how various instruments' items could be linked with factors that could be mapped to our operational definition. Our aim was to discern the extent to which this dimension was captured in the various instruments, whether or not the instruments' developers had envisaged these as representing respectfulness. Such an analysis of items relating to the concept of respectfulness is crucial given the emphasis in current PHC reforms on better addressing non-clinical expectations of patients and the way care is provided.

Method

The main methodological aspects have been described in detail elsewhere (Haggerty et al. 2011). Briefly: six instruments that evaluate PHC from the patient's perspective were administered to 645 healthcare users, balanced by English/French language, rural/urban location, low/high education and poor/average/excellent overall PHC experience.

Among the six instruments, two instruments comprised three subscales that mapped to our operational definition of respectfulness. The Primary Care Assessment Survey (PCAS) (Safran et al. 1998) had one, Interpersonal Treatment. The Interpersonal Processes of Care – Version II (IPC-II) had two: Interpersonal Style (Compassionate, Respectful) and Interpersonal Style (Disrespectful Office Staff). However, the IPC-II also had a Hurried Communication subscale that comprised three items out of five mapping with respectfulness. Additionally, there were individual items that assessed aspects of respectfulness in subscales measuring other attributes, notably in the Components of Primary Care Index (CPCI) (Flocke 1997), in the first version of the European general practice instrument (EUROPEP-I) (Grol et al. 2000) and in other subscales of the IPC-II (Stewart et al. 1999). No items or subscales directly addressed the provision of privacy, which was part of the operational definition and can be defined as the ability to seclude and reveal oneself selectively. The EUROPEP-I instrument addresses the notion of confidentiality of personal information. There was also considerable conceptual overlap with the concepts of interpersonal communication skills (ability to elicit and understand the patient's concerns) and trust. Although items from the PCAS Trust subscale were initially included in the respectfulness analyses, these items were later excluded as reflecting an outcome of care rather than an attribute of primary healthcare per se. In all, we retained 25 items (including four complete subscales) for analysis.

Analytic strategy

The analysis consisted of examining the distributional statistics and subscale correlations, followed by common factor and confirmatory factor analysis to identify dimensions common to the entire set of items, as outlined in detail elsewhere (Santor et al. 2011). Our strategy for analyzing respectfulness differed from that used for other attributes of care in this study. We departed from our overall strategy of honouring the subscales that had been validated by the instrument developers and included individual items from other subscales that mapped to our operational definition. This approach was necessary given that the creators of the measurement instruments did not identify respectfulness as a concept in their instruments but rather integrated various items that are conceptually linked with the operational definition of respectfulness that our experts identified.

Results

The item content and distribution of the responses are summarized in Table 1. Most of the individual items we mapped to respectfulness came from subscales related to interpersonal communication. Despite our design that specifically oversampled persons with a negative experience of care, the vast majority of respondents select the two most positive response options and very few the negative options. This is a well-known bias in the evaluation of experience of care by patients and may compromise the performance of exploratory factor analysis based on ordinary least squares regression techniques.

Table 2 presents the Pearson correlations among the three subscales that had been identified to capture respectfulness, as well as for IPC-II Hurried Communication, where three of the five items mapped to respectfulness. The correlations show that while subscales measuring physician respectfulness are highly correlated among themselves, these subscales do not correlate highly with the subscale for office staff respectfulness, suggesting that these are distinct constructs. To explore the extent to which respectfulness was distinct from or similar to other attributes, we calculated the mean of the correlations between each respectfulness subscale and the subscales from the other attributes. We see strong correlations with interpersonal communication. This could either be a true correlation between distinct constructs (two constructs that tend empirically to move in similar directions) or suggest a conceptual overlap between respectfulness and interpersonal communication (not truly distinct constructs).

Do all items measure a single attribute?

Using the subscales and items, we performed factor analysis with items mapping conceptually to respectfulness. We excluded from factor analysis all respondents with any missing values (listwise missing). This approach reduced our effective sample size from 645 to 519. Those excluded from the factor analyses were more likely than those included to be older and to have a clinic as the regular provider. Because this conservative approach can introduce bias, we repeated all the factor analyses using maximum likelihood imputation of missing values (Rubin 1987, Jöreskog and Sörbom 1996) to examine the robustness of our conclusions. Imputation increased our sample size slightly and improved the fit statistics in the confirmatory factor analysis models, but it did not change the magnitude of loadings nor the direction of our conclusions. This finding suggests that our conservative approach resulted principally in loss of statistical power rather than bias.

Although most of the items loaded reasonably well (loading >.40; see Table 3) onto a single latent variable using common factor analysis, the scree plots suggested a three-factor model. Based on our operational definition, we judged that the first factor (eigenvalue=13.99) is a reflection of "physician's interpersonal treatment" (interpersonal treatment), the second factor (eigenvalue=2.13), "interpersonal treatment by office staff" (office staff) and the third factor (eigenvalue=1.16) is related to "respect for the dignity of the person" (dignity). The EUROPEP-I rating of confidential treatment of the medical record loaded only moderately on interpersonal treatment, and the research team decided that it related conceptually to respect for the dignity of the person.

The dimensions of interpersonal treatment and dignity are highly correlated (.91), but office staff has much lower correlations with these two dimensions (.70 and .85, respectively). Structural equation modelling confirmed the three-factor structure of the data. The model and loadings of items are presented in Figure 1.

Discussion

We found three subscales and many items from available instruments that seemed to measure respectfulness and fit three elements in our operational definition: response to expectations in interpersonal treatment; respect for dignity; and provision of privacy by both physicians and clinic staff. Our analyses show that 18 of the 25 identified items load on one factor (interpersonal treatment). Indeed, the various experience-of-care constructs seem to correlate, and this correlation may emerge from a common latent construct. However, respectfulness by the staff is a separate construct or dimension that could come from a different reference point (e.g., the office staff rather than the physician). Finally, although our conceptual mapping placed wait time in the waiting room as a measure of respectfulness, our analysis suggests that this is probably more an indicator of accessibility, behaving very differently in factorial analyses.

Interpersonal treatment as a measure of respectfulness

The fact that a common factor seems captured by various items deserves further discussion. Many of these items, such as interpersonal communication and relational continuity, were developed for other constructs. This raises the possibility that experience of care in general relates to many different latent constructs, and that the same set of variables could capture different correlated constructs. This overlap could be related to the theoretical perspective adopted to subdivide the overarching concept of experience of care. In fact, the analysis of the face validity of respectfulness items suggest that healthcare providers can show their respectfulness through actions related to other attributes such as communicating well, taking time with patients, not making them wait and considering various aspects of their personality. In other words, for patients to feel respected, providers somehow have to translate their respect into actions related to communication, continuity, accessibility and comprehensiveness. Put this way, respect could be a determinant or marker of other aspects of experience of care of patients.

Interpersonal treatment and interpersonal style of communication could be part of the same construct. Given that communication and respect are closely related in the clinical encounter, communication is probably the best way to show respect, and miscommunication could generate mistrust and feelings that one is not respected or perceived as important.

Measurement implications

Our results suggest that interpersonal aspects of care have fuzzy boundaries and are probably difficult for people to distinguish. Overall, "my doctor is good and caring" might be the blanket statement that covers communication skills, relational quality and respect for the patient. The implication for measurement seems to be that dimensions related to the level of affiliation between patient and doctor are fraught with halo effects, a fact that probably impedes the clear-cut separation of concepts. Is the doctor disrespectful or a poor communicator? Is the doctor appropriate but cold? This finding might advocate for a measure of overall responsiveness, as suggested by the WHO in its framework of performance assessment, rather than the more specific concept of respectfulness. Still, such a broad concept remains difficult to measure and to link with specific improvement activities.

More recently, patient-centred care has been suggested as one essential quality of PHC and has been defined as encompassing aspects of providing care that is respectful and responsive to individual preferences, thus combining into a single concept the notion of respectfulness and responsiveness (Institute of Medicine 2001). This concept can also be useful in assessing notions of privacy and dignity, which do not seem to be captured by current instruments that assess respectfulness. This lack could be the object of further development of indices measuring patient-centred care, thus going beyond respectfulness while retaining more focus than with the broad, non-specific concept of responsiveness.

Respectfulness and quality of primary care

Our analyses suggest that respectfulness captures a wide array of the interpersonal aspects of care, and we found that among all the attributes measured, the respectfulness subscales discriminate most strongly between poor, average and excellent overall experience of care (Haggerty et al. 2011). This finding militates for a renewed interest in measuring respectfulness in healthcare evaluations.

The various Commonwealth Fund international healthcare policy surveys have highlighted that a fair proportion of the public seem to have lost confidence in the healthcare system and that what makes the difference in patients' satisfaction is more the way that care is provided rather than what care is provided (Schoen et al. 2004). Therefore, measurement of respectfulness and efforts to improve the aspects of care related to it might be levers to improve the performance of PHC care systems. This could complement the focus on medically necessary aspects of care and provide a more holistic approach to quality.

Limitations and strengths

We limited our analyses to validated instruments available in the public domain. Other instruments with various levels of validation or previous utilization exist as well and could capture some aspects related to the concept of respectfulness, such as the QUOTE questionnaire (Groenewegen et al. 2005; Kerssens et al. 2004) and the Commonwealth Fund Health Care Quality Survey (Blanchard and Lurie 2004).

We did not intend to assess respectfulness as a concept from the start. It was suggested by the operational definitions developed by the experts. These analyses represent a first assessment of the coverage and properties of items and subscales that could capture respectfulness in PHC settings. Further studies should aim at better evaluating tools and instruments for their capacity to evaluate respectfulness, develop more refined measures and assess the relevance of the broader concept of responsiveness.

Conclusion

This paper assessed how various items in current primary care experience instruments capture and relate to aspects of respectfulness. This is the first time a respectfulness lens has been applied to the most commonly used instruments. Our results show that most items related to the interpersonal aspects of treatment, while dignity and privacy are addressed poorly. However, our exploratory analysis provides guidance to evaluators to examine respectfulness in validated subscales. In the context of healthcare systems under economic stress and budgetary constraints, improvement in respectfulness could go a long way.

Le respect du point de vue du patient : comparaison entre instruments d'évaluation des soins de santé primaires

Résumé

Le respect est un élément central, et mesurable, de la réactivité des services de santé. La définition opérationnelle du respect est «la mesure dans laquelle les professionnels de la santé et le personnel de soutien satisfont aux attentes des patients en matière de traitement interpersonnel et font preuve de respect pour la dignité et l'intimité des patients.»

Objectif: Examiner dans quelle mesure le respect est capté par les instruments validés qui servent à évaluer les soins de santé primaires du point de vue du patient; et voir si les concepteurs ont envisagé, ou non, la notion du respect dans leur conception.

Méthode: Six cent quarante-cinq adultes, qui ont vécu au moins un contact avec leur clinique ou médecin régulier au cours des 12 mois antérieurs, ont répondu à six instruments d'évaluation des soins primaires, notamment deux sous-échelles qui touchent au respect : Interpersonal Processes of Care, version II (IPC-II, deux sous-échelles) et le Primary Care Assessment Survey (PCAS). De plus, le Components of Primary Care Index (CPCI) et la première version de l'EUROPEP (EUROPEP-I) comprennent des items liés au respect dans des sous-échelles qui servent à mesurer d'autres caractéristiques. Les résultats ont été normalisés pour permettre des comparaisons descriptives. Les analyses factorielles exploratoires et confirmatoires (modélisation par équation structurelle) ont servi à examiner l'adéquation à la définition opérationnelle.

Résultats: Les échelles liées au respect sont en étroite corrélation entre elles et avec la communication interpersonnelle. Tous les items présentent un point de saturation adéquat pour un facteur unique, qui est probablement le respect, mais le modèle qui s'ajuste le mieux comprend trois facteurs sous-jacents qui correspondent à (1) le traitement interpersonnel de la part du médecin (valeur propre = 13,99), (2) le traitement interpersonnel de la part du personnel de bureau (valeur propre = 2,13) et (3) le respect de la dignité (valeur propre = 1,16). La plupart des items captent le traitement interpersonnel de la part du médecin («style interpersonnel respectueux et avec compassion» et «communication hâtive» de l'IPC-II; «traitement interpersonnel» du PCAS). Le «style interpersonnel» de l'IPC-II (personnel de bureau irrespectueux) capte le traitement de la part du personnel, mais seulement trois items captent les aspects touchant à la dignité.

Conclusion: Dans les instruments validés disponibles, plusieurs items ou sous-échelles semblent mesurer la question du respect. Cependant, plusieurs de ces items sont reliés à d'autres construits tels que la communication interpersonnelle. Des recherches plus poussées devraient viser la conception de mesures plus précises, particulièrement pour ce qui est de l'intimité et de la dignité, et l'évaluation de la pertinence du concept général de réactivité.

Comments

Be the first to comment on this!