Skip Navigation


International Journal for Quality in Health Care Advance Access originally published online on July 17, 2008
International Journal for Quality in Health Care 2008 20(5):314-323; doi:10.1093/intqhc/mzn030
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
20/5/314    most recent
mzn030v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Hadjistavropoulos, H.
Right arrow Articles by Janzen, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hadjistavropoulos, H.
Right arrow Articles by Janzen, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press in association with the International Society for Quality in Health Care; all rights reserved

Patient perceptions of hospital discharge: reliability and validity of a Patient Continuity of Care Questionnaire

Heather Hadjistavropoulos1, Henry Biem2, Donald Sharpe1, Michelle Bourgault-Fagnou1 and Jennifer Janzen1

1 Department of Psychology, University of Regina, Regina, Saskatchewan, Canada
2 Department of Medicine, Lakeshore General Hospital, Pointe-Claire, Quebec, Canada

Address reprint requests to: Heather Hadjistavropoulos, E-mail: hadjista{at}uregina.ca


   Abstract
Top
 Abstract
Introduction
 Method
 Results
 Discussion
 Conclusion
 Funding
 Supplementary material
 Acknowledgements
 References
 
Background. Review of the literature reveals a need to develop a questionnaire that measures patient perceptions of factors impacting continuity of care following discharge from hospital. Such a measure has the potential to guide quality improvement initiatives related to continuity of care.

Objective. Our objective was to develop and examine the psychometric properties of a measure that would meet this need, the Patient Continuity of Care Questionnaire (PCCQ).

Method. The PCCQ was administered to 204 inpatients 4 weeks after discharge. The questionnaire was assessed by item and principal components analysis. Factors derived from principal components analysis were assessed for internal consistency and construct validity.

Results. A principal components analysis resulted in six subscales including perceptions of: (1) relationships with providers in hospital, (2) information transfer to patients, (3) relationships with providers in community, (4) management of written forms, (5) management of follow-up and (6) management of communication among providers. These subscales were internally consistent in our sample and demonstrated construct validity through correlations with other related constructs.

Conclusion. This initial study supports the reliability and validity of the PCCQ for measuring patient perceptions of factors central to continuity of care. The questionnaire subscales correspond to the theoretical components of continuity of care that have been proposed in the literature, namely informational, relational and management continuity. The subscales may be of value for identifying problems in continuity of care and for evaluating interventions aimed at improving continuity of care for patients after hospital discharge.

Keywords: continuity of care, hospital discharge, information, provider relationships, management, hospital processes, questionnaire, psychometrics


   Introduction
Top
 Abstract
 Introduction
Method
 Results
 Discussion
 Conclusion
 Funding
 Supplementary material
 Acknowledgements
 References
 
Continuity of care from one health-care setting to the next is a major concern as health-care systems become increasingly complex and fragmented. When transitions in care occur, gaps in care are common with the potential for morbidity and mortality [1, 2]. Haggerty et al. [3] conducted a comprehensive review of multidisciplinary research in the area of continuity of care. In their review, they distinguished among three forms of continuity of care. Informational continuity refers to the use of information (medical and personal) to provide appropriate client care over time, relational continuity refers to the consistency and quality of relationships between clients and providers as a means to connect care over time and management continuity refers to the provision of timely, coordinated and complementary services responsive to client needs connecting care over time.

Despite an improved understanding of and consensus that continuity of care is essential to health-care quality [4], there are few questionanires available that capture patient perceptions of continuity of care. Patient questionnaires have the potential to identify areas where care processes can be significantly improved. As recipients of care, patients are in an ideal position to report on factors that could impact on whether care is continuous, smooth or connected over time. For instance, they can report on the quality of relationships with providers that would encourage them to continue with care. They can also observe whether information was passed on from setting to setting or provider to provider. Further, they can report on the management of health-care documents, follow-up appointments and their observations about consistency and quality of communication among various providers. All of these dimensions may be best judged by the patient. Provider attempts to connect care over time and prepare the patient for discharge may be incompletely understood without garnering the patient's perspective.

Previously, we developed a measure of continuity of care, the Heart Continuity of Care Questionnaire, to allow for systematic examination of cardiac patient perceptions of varying aspects of continuity of care [5, 6]. The questionnaire was developed through review of the literature and refined by interviews with patients and providers. Based on this feedback, the questionnaire included items on perceptions of: (a) Information on Heart Condition, (b) Information on Medications, (c) Information on Physical and Dietary Needs, (d) Communication Among Providers, (e) Preparation for Discharge, (f) Post-hospital Care, (g) Post-hospital Review of Treatment and (h) Receipt of Conflicting Information. In the preliminary research on the measure, the questionnaire was found to provide reliable and useful information about patient care among individuals previously hospitalized and diagnosed with congestive heart failure and atrial fibrillation [6]. Follow-up study of the questionnaire revealed that the items addressed three general areas of care corresponding to informational, relational and management continuity. The scales had good reliability (subscales {alpha} ranged from 0.80 to 0.93) and demonstrated known groups validity in that continuity of care scores were lower for patients with more severe medical symptoms.

In this study, our aim was to adapt this questionnaire to be used with other patient populations since continuity of care is also relevant for patients with other health conditions. This paper describes the development of a broader Patient Continuity of Care Questionnaire (PCCQ) and tests of the reliability and validity of this measure in two different patient groups. We studied Orthopaedic inpatients, who had multiple disciplines involved in their care, and Family Medicine inpatients, who were older, suffered from multiple comorbidities and experienced longer lengths of stay due in part to difficulties in discharging patients to the care of community providers. This research was approved by the University of Regina Research Ethics Board and the Regina Qu'Appelle Health Region Ethics Board.


   Method
Top
 Abstract
 Introduction
 Method
Results
 Discussion
 Conclusion
 Funding
 Supplementary material
 Acknowledgements
 References
 
Participants
A total of 204 patients discharged from either an Orthopaedics unit at one hospital or a Family Medicine unit at a second hospital participated in the study. Participants were eligible to participate if they were over the age of 18; participants were excluded if in need of palliative care, cancer clinic care, transplant care, chronic renal failure care or psychiatric care. These patients were excluded since they are followed by the same team of providers over time, thus not allowing for the study of continuity from hospital providers to community providers.

Measures
The PCCQ (Appendix 1, online supplementary material) was developed by revising the Heart Continuity of Care Questionnaire. This questionnaire includes questions about the provision of information about the patient's heart condition, medications, precautions related to diet and exercise, discharge planning, post-hospital care, post-hospital review of treatment, receipt of conflicting information and communication among providers.

In adapting the Heart Continuity of Care Questionnaire to the PCCQ, modifications were made based on feedback from 29 experts across Canada. These experts were asked to provide us with input on continuity of care if they: (1) had published research on continuity of care or (2) were identified by a health-care administrator as having relevant expertise. The individuals resided in various Canadian provinces and had a variety of backgrounds: administrators/managers (n = 17), nurses (n = 4), physicians (n = 3) and researchers (n = 5). Individuals were asked what factors they felt were essential to continuity of care. This information was then used to modify the Heart Continuity of Care Questionnaire. The wording of the items was made more general so that the questionnaire would be applicable to multiple health conditions. The questionnaire was also separated into two sections to address the aspects of care prior to discharge (27 items) and following discharge (14 items). Some items on the Heart Continuity of Care Questionnaire were condensed into fewer items (e.g. four items that addressed information on medications were condensed into one item, four items addressing receipt of conflicting information were condensed into one item, seven items addressing post-hospital review of treatment were condensed into two items). In addition, to better address relationships with providers, nine items addressing relational continuity were added to assess patient perceptions of the relationship they had with their health-care providers both within the hospital and the community, including satisfaction with emotional support, information and opportunity to talk and raise questions, as well as confidence in providers, and a sense of being understood and feeling known by providers. Other items added addressed various support services/materials (four items), the involvement of informal caregivers in care (three items) and the completion and distribution of necessary forms (three items).

Like the Heart Continuity of Care Questionnaire, each item on the PCCQ was rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), as well as the option to choose not applicable. A respondent who did not receive services following discharge would choose the not applicable option.

In addition to the PCCQ, patients also answered questions about their background (i.e. age, gender, marital status, education, work status and whether they were residents of the city where the hospital was located). They also completed the Patient Reactions Assessment [7], a measure which assesses client satisfaction with physicians using three scales: (a) the Patient Information Index, which measures client satisfaction with the information given by the physician; (b) the Patient Affective Index, which measures client satisfaction with the emotional support from the physician and (c) the Patient Communication Index, which measures client satisfaction with the communication abilities of the physician. Alpha coefficients for the scales were as follows: 0.85, 0.85 and 0.82, respectively.

Patients also completed the Short Form-8 Health Survey (SF-8) [8], which results in summary measures of physical and mental health, with higher scores indicating better functioning and well-being. The SF-8 is reported to be both reliable and valid [8].

Procedure
A staff member from each unit asked eligible patients if they would participate. Those patients who consented were interviewed and received the questionnaires in their home approximately 4 weeks after discharge. In some cases, family members or personal guardians (nine from Family Medicine and one from Orthopaedics) assisted in providing responses for patients who desired such support. Patients were also asked for permission to have a nurse review their charts for information on length of stay, whether they received home care upon discharge, whether they received educational materials and whether they had been readmitted within 4 weeks of discharge.

Hypotheses
It was hypothesized that the factor structure of the PCCQ would resemble a recent conceptualization of continuity of care and has scales assessing the informational, relational and management aspects of continuity [3]. Construct validity of the PCCQ was expected to be demonstrated through moderate positive associations with the Patient Reactions Assessment [7], a measure of satisfaction with relationships with physicians. We also anticipated that the PCCQ might be related to SF-8 [8] subscales, but these correlations were expected to be weak, since the PCCQ is a measure of care provision and the SF-8 measures health status. Furthermore, we expected that the PCCQ scores would differ as a function of certain patient variables. We expected scores to be lower for patients who had more complex conditions (e.g. Family Medicine inpatients hospitalized for management of complex medical conditions compared with Orthopaedic surgical inpatients). Furthermore, we expected that the ratings would be lower for those with longer lengths of stay, since longer lengths of stay are typically a result of more complex conditions.

Statistical analyses
Analyses were undertaken using SPSS 11.0. Item analysis involved the examination of completion rates, means and standard deviations. We also calculated the percentage of patients expressing disagreement with an item (e.g. strongly disagree or disagree). Principal components analysis was used to examine the factor structure of the measure. Subscales resulting from this analysis were assessed for reliability using Cronbach's {alpha}. Pearson's product–moment correlations were calculated for continuous variables. Similar to previous authors [9], we defined a strong correlation as >0.70, moderate to substantial as 0.30–0.70 and weak as <0.30. Because of the many hypotheses tested, we only attributed statistical significance to P ≤ 0.01. Analysis of variance was used to examine whether subscale scores differed as a function of various patient characteristics (i.e. male versus female, married versus non-married, high school completion or not, retired or non-retired, urban versus rural residence, Orthopaedics versus Family Medicine unit, home care or not upon discharge).


   Results
Top
 Abstract
 Introduction
 Method
 Results
Discussion
 Conclusion
 Funding
 Supplementary material
 Acknowledgements
 References
 
Patient characteristics
Participants averaged 64.9 years of age (SD = 17.4). The majority were female (n = 122; 59.8%), and most were also married/common law (n = 116; 56.9%). The majority had a high school diploma or higher (n = 133; 66.7%). With regard to employment, most were retired (n = 118; 57.8%). Patients were hospitalized for management of a variety of conditions. Half of the patients (n = 102) resided in the city where the hospital was located, whereas the other half resided outside of the city. In the Family Medicine unit (n = 50), the primary diagnoses were: respiratory or pulmonary (n = 13; 26%), infection (n = 9; 18%), orthopaedic (n = 7; 14%), cardiovascular (n = 6; 12%), gastroenterological (n = 6; 12%), haematological (n = 3; 6%), urological (n = 2; 4%), chronic disease (n = 2; 4%) or dehydration (n = 2; 4%). The average length of stay in hospital was 19.10 days (SD = 9.53; range 4–63). Most participants were discharged directly to their homes (n = 39; 78%), with the remainder being discharged to a location providing support (i.e. convalescent care, personal-care home, apartment complex with supports). Approximately half (n = 28; 56%) of Family Medicine participants received home care upon discharge. Among the Orthopaedic participants (n = 154), hospitalization occurred for the following reasons: fracture (n = 50; 32.5%), knee replacement (n = 49; 31.8%), hip replacement (n = 36; 23.4%), back (n = 8; 5.2%), knee (n = 6; 3.9%), foot (n = 3; 1.9%), or elbow (n = 2; 1.3%) surgery. The average duration of hospital stay was 8.27 days (SD = 5.15). The majority of participants were discharged directly home from the hospital (n = 141; 91.3%) with only a small number (n = 31; 20.1%) receiving home care. Only 10% (n = 21) of all patients had no documentation of educational materials on their chart. Only 4% (n = 9) of patients were readmitted within 4 weeks of discharge from hospital for the same condition.

Item analysis
Table 1 reproduces the original 41 items of the PCCQ with item statistics. Twenty-seven items were answered by ~95% of the patients (i.e. 193 out of 204 participants gave a response other than not applicable). Over 50% of patients did not feel they or their caregiver had been informed of patient resources/supports and educational materials that would help ensure care was continuous. Over 30% of patients reported that upon discharge they were not provided with key information to help them manage their care after discharge (e.g. they did not feel they were told about non-urgent symptoms and how to cope with these, were not given information about urgent symptoms and whom to contact for these symptoms, were not told about clinical findings that could impact their future health care or were not given dietary instructions to manage their health condition). Similarly, over 30% of patients reported that their treatment had not been reviewed by their physician since they were discharged. Approximately 20% of patients felt that they had not been provided with information on their prognosis, restrictions in activities or treatment requirements after discharge. As many as 20% of patients reported that a concrete follow-up plan had not been developed or that they (or their caregivers) were not involved in the preparation of the follow-up plan. Lastly, 20% of patients reported that they had not been reminded of important appointments.


View this table:
[in this window]
[in a new window]

  		Table 1 Means, standard deviations and percentage of patients (n = 204) disagreeing with PCCQ items

 
Factor analysis
To examine the factor structure of the scale, we used only those items completed by ~95% of participants (n = 193). One exception to this exclusion criterion was that we included an item concerning medication (item 6: given information on medications) even though it was only completed by 186 participants. Given the importance of medication management to continuity of care, it was felt that this item should not be excluded from the analysis.

Kaiser's measure of sampling adequacy was 0.83, indicating that the 27 PCCQ items were appropriate for principal components analysis. The eigenvalues greater than 1 were as follows: 8.74, 2.59, 2.23, 1.74, 1.35 and 1.19. Since the eigenvalue greater than 1 criterion typically overestimates the number of factors to be analysed [10], we also performed parallel analysis [11, 12] using both mean and the 95th percentile eigenvalues to determine the appropriate number of factors to retain. This analysis supported a six-factor solution. These six factors accounted for 66.05% of the variance prior to rotation, with factor 1 accounting for 32.36% of the variance, and subsequent factors accounting for 9.61, 8.28, 6.44, 4.99 and 4.39% of the variance, respectively. The magnitude of the communalities indicated that the factors accounted for a substantial proportion of variance in the majority of items.

Examination of the simple structure following oblique rotation (promax) revealed there were no items that cross-loaded on more than one factor with a factor loading of greater than 0.40 (Table 2). This is positive in that items that load on more than one factor suggest a poor factor structure. There were two hyperplane items, which refer to items that do not load on any factors (‘All providers gave me consistent information about my care’ and ‘I was provided with information on restrictions in activities’). All factors had at least three items with loadings greater than 0.40. Given the apparent appropriateness of the six-factor solution, factor content was examined by attending to items that had salient loadings that were greater to or equal to 0.40 on a factor. The factors were interpreted as satisfaction with: (1) Relationships in Hospital, (2) Information to Patient, (3) Relationships in Community, (4) Management of Healthcare Forms, (5) Management of Follow-up and (6) Management of Communication Among Providers.


View this table:
[in this window]
[in a new window]

  		Table 2 Factor analysis of the PCCQ

 
Subscale analyses
The six-factor solution was further supported by adequate alpha coefficients of 0.88, 0.82, 0.85, 0.76, 0.71 and 0.68 for each factor, respectively. To further assess the PCCQ, we assessed the inter-correlations among the PCCQ subscales (Table 3). PCCQ subscale scores were created by calculating the average scale scores on the pertinent items. Generally, moderate correlations were found among the PCCQ subscales. The strongest correlations were found among the Relationship in Hospitals and Information Transfer and Management of Communication subscales (r = 0.52). Only two correlations were below r = 0.30: Information to Patient and Management of Communication Among Providers subscales correlated r = 0.28 and Relationships in Community and Management of Forms correlated r = 0.23.


View this table:
[in this window]
[in a new window]

  		Table 3 Correlations among PCCQ subscales

 
We explored whether there were statistically significant differences among subscale mean scores using repeated measures analysis of variance; a statistically significant effect was found, P < 0.0001. Simple effects analyses showed that Relationships in Hospital (M = 4.41, SD = 0.84), Relationships in Community (M = 4.46, SD = 0.72) and Management of Forms (M = 4.64, SD = 0. 69) were rated higher on the 1–5 scale than Information Transfer (M = 4.00, SD = 1.07), Management of Follow-up (M = 4.13, SD = 1.11) and Management of Communication among Providers (M = 4.11, SD = 1.02).

Correlations
To further assess the PCCQ, we calculated the correlation coefficients among the PCCQ subscales and other continuous variables measured in the study (Table 4). Supporting the convergent validity, the PCCQ subscales were correlated with the PRA subscales which captures the ratings of the patient–physician relationship. The correlations were generally moderately strong. All relationships among PCCQ and Patient Reactions Assessment subscales were statistically significant with only a few exceptions. The Management of Forms subscale of the PCCQ was not correlated with the measures of physician affect or physician communication. Also the PCCQ Management of Communication Among Providers was not correlated with the ratings of physician communication.


View this table:
[in this window]
[in a new window]

  		Table 4 Correlations among PCCQ subscales and continuous measures of patient–physician relationship, SF-8 physical and mental health, age and length of stay

 
Supporting the divergent validity of the PCCQ, the Physical subscale of the SF-8 was not related to the PCCQ subscales. The Mental Health Subscale of the SF-8 was correlated only with Relationships in Hospital and Information Transfer scores. These correlations were generally moderate showing that the greater the perceptions of continuity of care in these areas, the greater the self-rating of mental health.

Age and length of stay, continuous background variables, were also examined for their relationships with the PCCQ subscale scores. Younger patients had higher scores on the Information Transfer subscale. Length of stay was significantly correlated with several PCCQ subscales. Patients with shorter lengths of stay had higher scores on subscales measuring Relationships in Hospital, Information Transfer and Management of Follow-up.

Group comparisons
Analysis of variance was used to examine whether different patient groups differed in PCCQ subscale scores (Table 5). PCCQ subscales scores were not significantly different among males and females, those with or without a high school education or those who were or were not retired.


View this table:
[in this window]
[in a new window]

  		Table 5 Mean PCCQ subscale scores by patient groups

 
A number of patient groups differed in their ratings on the Information subscale. Married participants differed from non-married participants in ratings of Information, with married individuals giving higher ratings to the Information (P < 0.01). Individuals residing in the city compared to those residing outside of the city gave lower ratings to Information (P < 0.01) as well as Relationships in the Community (P < 0.01). Orthopaedic patients compared with Family Medicine patients gave higher ratings to Information (P < 0.01) and also Management of Follow-up (P < 0.001). Finally, patients who received home care after discharge compared with those who did not gave lower ratings to Information (P < 0.01).


   Discussion
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
Conclusion
 Funding
 Supplementary material
 Acknowledgements
 References
 
This initial study of the PCCQ suggests that the questionnaire has considerable potential for capturing patients' perceptions of factors related to continuity of care. Item analysis revealed that by examining the 41 items, it was possible to identify areas that could disrupt the continuity of care. Providers interested in improving continuity of care could easily use item analysis of the questionnaire to identify areas for improvement of care related to continuity of care. Using this method of analysis, in this particular study, we identified numerous areas for improvement. As an example, we found that over 50% of patients did not feel they or their caregiver had been informed of resources, supports and educational materials that would help ensure continuous care. Furthermore, as many as 30% of patients reported they were not provided with key information upon discharge to help with continuity of care (e.g. they were not told about non-urgent symptoms and how to cope with them, they had not been given information on urgent symptoms and whom to contact for these symptoms, they were not told about clinical findings that could impact their future health). It is likely that administering the questionnaire and reviewing the percentage of patients who disagree with items could help providers identify areas where improvements could be made to reduce gaps in continuity of care post-discharge.

Further analysis of the PCCQ revealed that 27 of the items were common to most patients who were studied and therefore could be used to develop subscales of continuity of care. The factor analysis of these items identified six factors. Examination of the content suggested that the factors could be labelled in a manner consistent with the three areas of continuity identified by Haggerty et al. [3]. They in their multidisciplinary review of the literature recognized that continuity of care is multifaceted and does not represent a single construct. Factors 1 and 3 had questions that pertained to relationships with providers in hospital (factor 1) and the community (factor 3) reflecting Relational Continuity. Factor 2 consisted of items concerning provision of information to patients in order to ensure care is continuous and thus pertained to Informational Continuity. Factors 4–6 reflected varying dimensions of Management Continuity, namely management of healthcare forms, management of follow-up appointments and management of communication among providers.

In this study, we created a mean subscale score for each factor. These mean subscale scores correspond to the 1–5 response scale, with higher scores reflecting more positive perceptions of care. The advantage of creating mean subscale scores is that it allows us to compare ratings on mean subscale scores even though each scale has a varying number of items. Examination of the mean scores on the factor scores in this study yielded interesting information. In particular, patients in this sample obtained higher scores on subscales reflecting Relationships within Hospital and Relationships in the Community as well as Management of Follow-up. They obtained significantly lower scores on scales reflecting Information Transfer, Management of Follow-up and Management of Communication Among Providers. Others who use the PCCQ in the future could potentially use the mean subscale scores to create a ‘continuity of care profile’, which could be used to identify general areas where improvements need to be made (e.g. Information Transfer, Management of Follow-up) and could also be tracked over time.

In terms of relationships among subscales, although we found that the subscales scores are correlated in this study, the pattern of correlations and the factor structure suggest that the measure is capturing distinct facets of continuity of care. Examination of the pattern of correlations suggests that some of the dimensions are more related than others. The strongest correlations, for instance, were found among subscales assessing the Relationships in Hospitals, Information Transfer and Management of Communication Among Providers. It would seem that patients feel more positively about providers in hospital when providers also do a good job providing patients with information and communicate consistently and effectively with each other.

In addition to providing information on the factor structure of the PCCQ, this study supported the internal consistency of the subscales and provided further information demonstrating the validity of the scales. For example, according to our hypotheses and consistent with convergent and divergent validity (e.g. similar measures correlate more strongly than divergent measures), ratings of information, affect and communication of the physician from the Patient Reactions Assessment measure correlated with the PCCQ subscales more strongly than measures of mental and physical health status.

Also of interest in this study was that Information subscale appeared to vary as a function of several patient variables. In particular, patients who were older, single, residing in the city, inpatients on Family Medicine Unit and provided with home care upon discharge gave lower ratings to the information they received compared with other patient groups. The findings are not causal in nature, but could reflect that care is less complicated among younger patients and information transfer is less of a concern when patients are married and have this additional support. We unexpectedly found that ratings were higher among patients residing outside of the city and those who did not receive home care; upon reflection, this may suggest that more attention is given to patients being transferred out of the city or when home care is not involved when it is evident that patients are at risk of discontinuity in care. Overall, findings such as these suggest that it may be possible to use the PCCQ scales to identify vulnerable groups who require further resources when it comes to transfer of information.

As noted in the introduction, the PCCQ was based on the Heart Continuity of Care Questionnaire [5, 6]. Although revisions were made to Heart Continuity of Care questionnaire to allow it to be used with other patient groups, it appears that the psychometric strengths of the questionnaire have been retained. The PCCQ like the HCCQ has evidence to suggest that the items address general areas of care corresponding to informational, relational and management continuity. In both cases, reliability and validity of the scales were supported. Furthermore, both measures appear to have practical value allowing for identification of areas of care that require improvement. More specially, with both measures, lower subscale or item scores may reflect areas that could be targeted for improvements in continuity of care.

Taken together the results of the present study suggest that the PCCQ captures patient perceptions of factors related to continuity of care. At this time, given that the PCCQ is still early in its development, we recommend that the full 41 items be administered. This full measure allows for a more systematic examination of the nature and extent of problems in continuity of care compared with the 27-item version. That said, if time is a constraint, then the 27-item version still provides considerable information. Both the 41-item and 27-item versions allow for the examination of subscale scores that are likely to be of significant value in studies evaluating interventions aimed at improving continuity of care.

This study has some limitations. The study sample was made up of patients discharged from an Orthopedics unit and a Family Medicine unit. Thus, the findings may not be applicable to other patient populations. On the other hand, these patients had a wide variety of medical primary diagnoses and co-morbid conditions. A volunteer effect may also have impacted our study results meaning that those agreeing to interviews may have been different from the target population. It is also important to note that the PCCQ is only appropriate for patients who demonstrate a certain degree of independence in their care. Other measures are needed for patients who have difficulties communicating (e.g. focused on caregiver perceptions). Finally, this represents a preliminary study of the PCCQ and a further study of the validity of the measure is needed.

In particular, in future studies, further information on the construct validity of the PCCQ should be collected. It would be helpful to examine how the PCCQ compares with another measure of continuity of care, such as the Care Continuity Index [13]. This index has 12 items and provides a brief indication of the patient's perceptions of continuity of care and therefore would represent an interesting comparison. Also of interest is further comparison of the PCCQ with measures of patient satisfaction, such as the Client Satisfaction Questionnaire [14], which is a well-validated instrument of service evaluation. It consists of eight questions about general satisfaction (e.g. how would you rate the quality of service you received). The PCCQ primarily asks about whether aspects of care took place (e.g. provision of information), but also has some questions addressing general satisfaction with care (e.g. satisfaction with emotional support), and therefore should likely correlate at least moderately with patient satisfaction measures.

The PCCQ also needs to be examined for generalizability and utility with other patient groups and in other organizations. Furthermore, it should be examined for its ability to detect problems and measure improvement over time. It may also be beneficial to study the PCCQ using other modes of administration. In this study, we chose to conduct in-person interviews, but it is also quite possible that information could be compiled over the Internet or by phone or mail. The optimal timeframe for evaluation of post-discharge continuity of care could also be examined. In this study, we used a 4-week timeframe; however, other timeframes (shorter or longer) could also be studied.


   Conclusion
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
Funding
 Supplementary material
 Acknowledgements
 References
 
The PCCQ was found to have adequate psychometric properties for measuring continuity of care. The PCCQ appears to encompass the theoretical components of relational, informational and management continuity. Further study is needed to support the validity, generalizability and utility of this instrument, but the current results suggest the PCCQ may be of value for identifying negative perceptions of factors that may interfere with continuity of care.


   Funding
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
 Funding
Supplementary material
 Acknowledgements
 References
 
This study was made possible by a grant (FRN 73183) funded by three partners: the Canadian Institute of Health Research, the Canada Health Services Research Foundation and Saskatchewan Learning. In-kind costs were also contributed by the Regina Qu'Appelle Health Region.


   Supplementary material
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
 Funding
 Supplementary material
Acknowledgements
 References
 
Supplementary material is available at International Journal for Quality in Health Care online.


   Acknowledgements
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
 Funding
 Supplementary material
 Acknowledgements
References
 
We are indebted to the patients, providers, health-care decision makers and research assistants who made this study possible.


   References
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
 Funding
 Supplementary material
 Acknowledgements
 References
 

  1. Moore C, Wisnivesky J, Williams S, McGinn T. Medical errors related to discontinuity of care from an inpatient to an outpatient setting. J Gen Intern Med (2003) 18:646–51.[CrossRef][Web of Science][Medline]

  2. van Walraven C, Mamdani M, Fang J, Austin PC. Continuity of care and patient outcomes after hospital discharge. J Gen Intern Med (2004) 19:624–31.[CrossRef][Web of Science][Medline]

  3. Haggerty JL, Reid RJ, Freeman GK, Starfield BH, Adair CE, McKendry R. Continuity of care: a multidisciplinary review. Br Med J (2003) 327:1219–21.[Free Full Text]

  4. Donaldson MS. Continuity of care: a reconceptualization. Med Care Res Rev (2001) 58:255–90.[Abstract/Free Full Text]

  5. Kowalyk K, Hadjistavropoulos HD, Biem HJ. Measuring continuity of care for cardiac patients: development of a patient self-report questionnaire. Can J Cardiol (2004) 20:205–12.[Web of Science][Medline]

  6. Hadjistavropoulos HD, Biem HJ, Kowalyk K. Measuring continuity of care in cardiac patients: reliability and validity of an in person questionnaire. Can J Cardiol (2004) 20:883–91.[Web of Science][Medline]

  7. Galassi JP, Schanberg R, Ware WB. The Patient Reactions Assessment: a brief measure of the quality of the patient-provider medical relationship. Psychol Assess (1992) 4:346–51.[CrossRef]

  8. Ware JE, Kosinski M, Dewey JE, Gandek B. How to Score and Interpret Single-item Health Status Measures: A Manual for Users of the SF-8 Health Survey. (2001) Lincoln, RI: QualityMetric Incorporated.

  9. McHorney CA, Ware JE Jr, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care (1993) 31:247–63.[Web of Science][Medline]

  10. Tabachnick BG, Fidell LS. Using Multivariate Statistics (1989) 2nd edn. New York: Harper & Row.

  11. Horn JL. A rationale and test for the number of factors in factor analysis. Psychometrika (1965) 30:179–85.[CrossRef][Web of Science][Medline]

  12. Longman RS, Cota AA, Holden RR, Fekken GC. A regression equation for the parallel analysis criterion in principal components analysis: mean and 95th percentile eigenvalues. Multivar Behav Res (1989) 24:59–69.[CrossRef]

  13. Bull MJ, Luo D, Maruyama GM. Measuring continuity of elders' posthospital care. J Nurs Meas (2000) 8:41–60.[Medline]

  14. Attkisson CC, Zwick R. The client satisfaction questionnaire. Eval Program Plann (1982) 5:233–7.[CrossRef][Medline]

Accepted for publication June 18, 2008.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
20/5/314    most recent
mzn030v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Hadjistavropoulos, H.
Right arrow Articles by Janzen, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hadjistavropoulos, H.
Right arrow Articles by Janzen, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?