Abstract
Background
Previous research using cross‐sectional data has shown a positive relationship between patient activation and quality of care. The quantitative relationships in the same patients over time, however, remain undefined.
Objective
To examine the relationship between changes in activation over time and patient‐assessed quality of chronic illness care.
Design
Prospective cohort study.
Participants
The study used data reported annually from 2008 (N = 3761) to 2010 (N = 3040), using self‐report survey questionnaires, completed by patients with type 2 diabetes in a population‐based cohort in Queensland, Australia.
Main Measures
Principal measures were the 13‐item Patient Activation Measure (PAM), and the 20‐item Patient Assessment of Chronic Illness Care (PACIC) instrument.
Methods
Nonparametric anova was used to determine the association between patient activation and patient‐assessed quality of care in low and high patient activation groups at baseline (2008), and in 2009 and 2010, when patients had changed group membership. The Wilcoxon signed ranks test was used to compare the PACIC scores between baseline and each follow‐up survey for the same patient activation level.
Results
Patient activation was positively associated with the median PACIC score within each survey year and within each of the groups defined at baseline (high‐ and low‐activation groups; P < 0.001).
Conclusions
Patient activation and the PACIC change in the same direction and should be considered together in the interpretation of patient care assessment. This can be carried out by interpreting PACIC scores within strata of PAM.
Keywords: chronic illness care, patient activation, Patient Assessment of Chronic Illness Care, patient experiences, quality assessment
Introduction
Application of the chronic care model (CCM) in diabetes care has been shown to improve care outcomes.1, 2 The CCM comprises six components, which are organization of health care, community resources, self‐management support, delivery system design, decision support and clinical information systems.3 Among these components, ‘self‐management support’ and ‘delivery system design’ directly involve patients in the care process, and hence, the quality of care regarding these components can be measured through patient perspectives. The Patient Assessment of Chronic Illness Care (PACIC) instrument is specifically designed to assess the extent to which care is congruent with these components of the CCM.4 The assessment would then reflect the extent to which patients believe their providers empower them in engaging or taking an active role in their health care.
The PACIC has been validated in various health‐care settings in different countries (including Australia), and for various chronic conditions, including diabetes. Rick and colleagues have summarized psychometrics data on the PACIC, which was found to be reliable and valid.5 Previous studies have shown that the PACIC was correlated with other quality of care measures, including care experience surveys and the receipt of laboratory tests and self‐management support.4, 6, 7, 8, 9, 10 However, its association with health outcomes such as glycaemic control and blood cholesterol levels remains equivocal.11, 12
While quality assessments from patient perspectives reflect the technical quality of providers (e.g. providers’ adherence to recommended care), it may also be influenced by the customer quality (attributes of patients that enable them to participate more effectively in the health‐care delivery system).13 Such patient attributes might be expected to influence patient assessment of the technical quality of care.13, 14 One important patient attribute is patient activation, which can be measured using the Patient Activation Measure (PAM) and is defined as the patients’ ability to take an active role in their health care.15 Patients’ intrinsic ability to self‐manage their care, such as being informed, confident and proactive, may shape how they use and rate the quality of health care. Thus, patients’ ratings of the care they receive may be different from the providers’ actual delivery of care.16
Previous research supports the role of patient activation in the assessment of quality of care.17, 18, 19, 20 Several studies using the PACIC measure also reported a positive correlation between quality of care and patient activation4, 21 or self‐management behaviours.9, 10 However, we did not identify any studies that directly explore the relationship between patient activation and quality of care using longitudinal data despite this being flagged previously. Our study thus aims to explore the association between patient activation and patient‐assessed quality of care in patients followed up at two points in time. The use of such follow‐up data in the same patients will demonstrate how changes in patient activation at each time point relate to their PACIC assessments.
Research design and methods
The study
The Living with Diabetes Study (LWDS) is a prospective cohort study, conducted annually in Queensland, Australia, from 2008 to 2011. Ethics approval for the study was gained from the University of Queensland's Behavioral and Social Sciences Ethical Review Committee. The self‐report survey questionnaires and consent forms were mailed to a sample of 14 439 adult registrants of the National Diabetes Services Scheme (NDSS), which covers an estimated 80–90% of the Australian population diagnosed with diabetes.22 Details on the study participants and overview of the measures have been reported previously.23 The response rate was 29% with completed questionnaires returned at baseline by 3951 respondents, who were diagnosed with either type 1 or type 2 diabetes. The respondents showed similar characteristics to non‐respondents, except that they were older and less likely to be of Indigenous origin.24 They were more likely to be male, 50–69 year old, non‐indigenous, recently registered in the NDSS and not using insulin, compared to 129 900 non‐study NDSS registrants in Queensland (either non‐consenting registrants or non‐sampled consenting registrants or non‐respondents).24 This study used data reported by respondents with type 2 diabetes only, in 2008 (N = 3761), 2009 (N = 3209) and 2010 (N = 3040).
Principal measures
The PACIC instrument measures the quality of chronic illness care from patient perspectives.4 It consists of 20 items on the chronic care the patients received, with each item scored using a five‐point scale, ranging from 1 (none of the time) to 5 (always). The overall PACIC score is obtained by averaging across all items answered (excluding items with missing data if any). Higher scores indicate perceptions of higher quality care. The developers of the PACIC, Glasgow et al. (2005), proposed five a priori subscales: patient activation, delivery system design, goal setting, problem‐solving and follow‐up/coordination. However, the factor structure of the PACIC has been highly debated, and previous studies which accounted for the ordinal nature of the PACIC (including the study using the LWDS data) recommended the use of the single‐factor structure.25 In addition, the PACIC focuses on the assessment of self‐management support as all subscales proposed by Glasgow et al. (2005) map mainly onto self‐management support component of the CCM. Thus, we used the PACIC as a single scale in this study.
To measure patient activation, we used the 13‐item PAM,26 which has been shown to be reliable and valid in various populations and settings.27, 28, 29, 30 Each PAM item bears a statement on the patient's knowledge, attitude, skills or confidence in self‐managing their health care and is scored using a four‐point scale, ranging from 1 (strongly disagree) to 4 (strongly agree) on the statement. Thus, the total raw score ranges from 13 to 52.15, 26 In the case of PAM items with missing data, the total score of the items answered was divided by the number of items answered, and then multiplied by 13 to get the raw score. The raw score was converted to an activation score ranging from 0 to 100 according to the scoring instructions for the PAM (2008). Each respondent was then assigned into one of four patient activation levels, with Level 1 being ‘least activated’ to Level 4 being ‘most activated’, using the following cut‐off points of PAM scores: ≤47 for Level 1; ≥47.1 and ≤55.1 for Level 2; ≥55.2 and ≤67 for Level 3; and ≥67.1 for Level 4.31
Statistical analyses
Descriptive analyses provide frequencies, proportions, means and standard deviations of respondent characteristics, and median and interquartile ranges (IQR) of principal measures for this study at baseline. The t‐test was used to examine mean differences in patient characteristics, and the chi‐square test was used to examine differences in proportions according to low‐ or high‐activation groups. Mean changes in the PACIC were determined for each change in patient activation level over time.
To determine how the PACIC overall score was associated with patient activation in each of the survey years (2008, 2009 and 2010) and change in activation over time (from 2008 to 2009 and from 2008 to 2010), we categorized the baseline PAM data (2008) into a low‐activation group (levels 1 and 2) and a high‐activation group (levels 3 and 4). Participants who did not respond to any of the PACIC or PAM items were excluded from the study. The PACIC scores and PAM scores for the rest of the participants were calculated as stated above (without imputation for missing data). A nonparametric Kruskal–Wallis analysis of variance (anova) was carried out by stratum, and graphs were plotted for group medians and 95% confidence intervals (CIs),32 both at baseline and on follow‐ups in 2009 and 2010. This analysis was repeated using (i) a subset of respondents reporting no change in activation level across all 3 years, (ii) data only for patients who participated in all three survey years (to account for attrition) and (iii) data for a sample who reported 10 or more PACIC items (to account for missing data).
For the baseline low‐activation group (levels 1 and 2), the baseline median PACIC score for each of the two patient activation levels was compared with the median PACIC score for the corresponding activation levels in subsequent years (using the Wilcoxon signed ranks test). The same analysis was repeated for the baseline high‐activation group (levels 3 and 4). Although this strategy may seem to suggest restriction to changes in patient activation in either direction, this is not the case because our analyses in 2009 and 2010 included all available participants regardless of whether change in their activation was upward or downward. We simply stratified patient activation at baseline such that the patients who increased activation over time were in one group and those who decreased activation in the other. Overall, however (i.e. combining both strata), patients could either increase or decrease patient activation and the stratification aimed to clarify demonstration of what happened to the PACIC when activation levels changed.
No confounders were considered as the PAM and the PACIC were the only variables of interest, and the purpose of this study was simply to determine whether these two measures are associated or not. It was not the purpose of this analysis to determine whether the association between the PAM and the PACIC (if any) was causal or not. Significance was considered at P < 0.05. MedCalc statistical software (MedCalc® Version 12.6.1.0, Ostend, Belgium) and SPSS statistical software (IBM® SPSS® Statistics 20.0, Chicago, IL, USA) were used for the analyses.
Results
At baseline, the mean age of the participants was 62.5 ± 10.9 years, the proportion of female respondents 45%, and that of Aboriginal and/or Torres Strait Islander respondents 1.8%. As expected, disease progression was evident with more respondents reporting they used insulin in 2009 (21%) and again in 2010 (22%) compared to 2008 (18%; P < 0.001). Half of this cohort (48%) had completed less than high school education with the remainder equally distributed among other ‘educational attainment’ categories, which were completion of high school, certificates/diplomas, trade/apprenticeship and university degrees.
The follow‐up rate from baseline to 2009 was 86% and from baseline to 2010 was 81%. Participants were more likely to respond to the survey at the follow‐up in 2009 if they were older, had a university degree, were married or lived in the inner regional area at the baseline. They were more likely to respond in 2010 if they were retired or married and less likely to respond if they lived in outer regional area and were using insulin at the baseline. The number of respondents in all 3 years was 2891.
Proportion of participants with missing data on PACIC items varied from 12.4 to 15.8% in 2008 (N = 3761), 13.1 to 15.3% in 2009 (N = 3209), and 9.1 to 10.4% in 2010 (N = 3040), depending on the items.25 This proportion varied from 0.8 to 2.1% in 2008, 0.7 to 2.1% in 2009, and 0.5 to 2.1% in 2010 when considering only participants who responded to 10 or more PACIC items in each year (N = 3222 in 2008; N = 2764 in 2009; and N = 2765 in 2010). On the other hand, 73.9% of all participants in 2008, 75.4% in 2009 and 80.4% in 2010 completed all 20 PACIC items. However, 8.6% of participants (n = 323) in 2008, 8.5% (n = 272) in 2009 and 5.3% (n = 161) in 2010 were excluded from the analyses as they did not respond to any of the 20 items.
In this study, at least 95% of all participants in any of the 3 years responded to all 13 PAM items. Among the participants with at least 1 PACIC item answered, 96.3% in 2008 (N = 3438), 97.4% in 2009 (N = 2937) and 97.0% in 2010 (N = 2879) responded to all 13 PAM items, while seven cases each in 2008 and 2009, and 10 cases in 2010 did not respond to any of the 13 items and thus further excluded from the study. Throughout the 3 years of study (2008–2010), 973 participants reported no change in activation level.
Differences in patient characteristics at baseline according to low and high patient activation levels are presented in Table 1. Although many of these differences were statistically significant, practically meaningful differences in proportions of patients in each of these activation levels were small (<10%), with the largest differences seen in educational attainment categories (5.6%) and employment categories (6.9%).
Table 1.
Patient characteristics at baseline according to low and high patient activation levels
| Patient activation level | P valuea | ||
|---|---|---|---|
| 1 and 2 | 3 and 4 | ||
| Mean age in years (SD) | 61.6 (11.51) | 62.8 (10.53) | 0.001 |
| Gender | 0.041 | ||
| Female | 481 (42.3) | 1190 (45.9) | |
| Male | 656 (57.7) | 1402 (54.1) | |
| Ethnicity | 0.637 | ||
| Aboriginal or Torres Strait Islander | 21 (1.9) | 42 (1.7) | |
| Others | 1084 (98.1) | 2462 (98.3) | |
| Education attainment | <0.001 | ||
| Less than high school | 572 (51.6) | 1159 (46.0) | |
| Completed high school | 150 (13.5) | 295 (11.7) | |
| Certificate/diploma | 135 (12.2) | 327 (13.0) | |
| Trade certificate/apprenticeship | 135 (12.2) | 380 (15.1) | |
| University Bachelor degree or higher | 117 (10.6) | 359 (14.2) | |
| Employment status | <0.001 | ||
| Full/part‐time/self‐employed | 411 (36.7) | 970 (38.0) | |
| Unemployed | 89 (7.9) | 219 (8.6) | |
| Retired | 480 (42.8) | 1216 (47.7) | |
| Unable to work | 141 (12.6) | 145 (5.7) | |
| Annual income | 0.041 | ||
| Less than $20 000 | 344 (33.8) | 680 (30.2) | |
| $20 000 or more | 673 (66.2) | 1568 (69.8) | |
| Marital status | 0.093 | ||
| Never married/ Widowed/ Divorced/ Separated | 350 (31.0) | 725 (28.3) | |
| Married/De facto | 779 (69.0) | 1839 (71.7) | |
| ARIA | 0.506 | ||
| Major cities | 736 (65.9) | 1631 (63.9) | |
| Inner regional | 239 (21.4) | 604 (23.7) | |
| Outer regional | 127 (11.4) | 281 (11.0) | |
| Remote | 15 (1.3) | 36 (1.4) | |
| BMI: Mean (SD) | 32.4 (7.23) | 30.7 (6.35) | <0.001 |
| Mean duration of diabetes in years (SD) | 7.9 (7.68) | 7.3 (8.00) | 0.041 |
| Treatment status | <0.001 | ||
| Insulin requiring | 259 (22.8) | 416 (16.0) | |
| Oral medications only | 689 (60.7) | 1564 (60.3) | |
| Diet and/or exercise or no/other treatment | 188 (16.5) | 612 (23.6) | |
| Mean number of complications (SD) | 1.4 (1.49) | 1.0 (1.26) | <0.001 |
| Mean number of comorbidities (SD) | 1.3 (1.47) | 1.0 (1.27) | <0.001 |
ARIA, Accessibility/Remoteness Index of Australia.
Data are presented in N (%) unless otherwise indicated.
T‐test for mean differences and chi‐square statistics for differences in proportions.
Overall, the median PAM (60; IQR 52.9–73.1) and median PACIC (2.2; IQR 1.5–3.3) scores were stable across the 3 years. Table 2 shows a positive mean change in PACIC scores for each upward change in patient activation level over time, and vice versa. Similarly, when patient activation level decreased or increased in subsequent years, the median PACIC scores moved in the same direction as seen in Figure 1. Findings from the Kruskal–Wallis anovas are presented in the footnotes for Figure 1 (P < 0.001 in all analyses). Additionally, the Wilcoxon signed ranks test showed that median PACIC scores between the baseline and each follow‐up survey did not differ for the same activation level as seen in Table 3 (P > 0.15 in all analyses). This finding was consistent through each of the two follow‐up survey years for both the baseline high‐ and the baseline low‐activation groups.
Table 2.
Mean change in the PACIC score from 2008 to 2009 and 2008 to 2010 by change in patient activation level
| Patient activation 2009 n (mean change in PACIC score) | Patient activation 2010 n (mean change in PACIC score) | |||||||
|---|---|---|---|---|---|---|---|---|
| Level 1 | Level 2 | Level 3 | Level 4 | Level 1 | Level 2 | Level 3 | Level 4 | |
|
Patient Activation 2008 n (mean change in PACIC score) | ||||||||
| Level 1 | 146 (0.10) | 119 (0.05) | 63 (0.18) | 25 (0.35) | 141 (0.07) | 89 (0.09) | 61 (0.18) | 35 (0.44) |
| Level 2 | 80 (−0.05) | 157 (−0.01) | 171 (0.15) | 82 (0.23) | 59 (−0.24) | 151 (0.05) | 163 (0.17) | 99 (0.08) |
| Level 3 | 55 (−0.13) | 145 (−0.10) | 491 (0.05) | 260 (0.05) | 60 (−0.29) | 154 (−0.15) | 454 (−0.04) | 273 (0.04) |
| Level 4 | 22 (−0.46) | 68 (−0.13) | 229 (−0.11) | 613 (0.01) | 24 (−0.16) | 54 (−0.45) | 234 (−0.32) | 612 (−0.02) |
| Total (n) | 303 | 489 | 954 | 980 | 284 | 448 | 912 | 1019 |
N (2008–2009 panel) = 2726; N (2008–2010 panel) = 2663.
Figure 1.
Overall PACIC score according to patient activation level. Error bars represent median PACIC score with 95% CI. (a): N = 2376; n (activation level 3) = 1202; n (activation level 4) = 1174; P < 0.001. (b) N = 2007; n (activation level 1) = 76; n (activation level 2) = 230; n (activation level 3) = 781; n (activation level 4) = 920; P < 0.001. Median PACIC scores were significantly different (P < 0.05) between any two activation levels, except between level 1 and level 2. (c) N = 1995; n (activation level 1) = 88; n (activation level 2) = 226; n (activation level 3) = 740; n (activation level 4) = 941; P < 0.001. Median PACIC scores were significantly different (P < 0.05) between any two activation levels, except between level 1 and level 2. (d) N = 1051; n (activation level 1) = 432; n (activation level 2) = 619; P < 0.001. (e) N = 896; n (activation level 1) = 233; n (activation level 2) = 304; n (activation level 3) = 247; n (activation level 4) = 112; P < 0.001. Median PACIC scores were significantly different (P < 0.05) between any two activation levels, except between level 1 and level 2 and between level 3 and 4. (f) N = 847; n (activation level 1) = 210; n (activation level 2) = 259; n (activation level 3) = 240; n (activation level 4) = 138; P < 0.001. Median PACIC scores were significantly different (P < 0.05) between any two activation levels.
Table 3.
Comparison of median PACIC overall scores between baseline and each follow‐up survey for the same patient activation level
| N | Median PACIC score | P value for comparison of PACIC score from baseline | |
|---|---|---|---|
| Baseline low‐activation group (activation levels 1 and 2) | |||
| Patient activation level 1 in 2008 (baseline) | 432 | 1.78 | |
| Patient activation level 1 in 2009 | 238 | 1.73 | 0.194 |
| Patient activation level 1 in 2010 | 217 | 1.65 | 0.199 |
| Patient activation level 2 in 2008 (baseline) | 619 | 2.00 | |
| Patient activation level 2 in 2009 | 299 | 1.95 | 0.701 |
| Patient activation level 2 in 2010 | 252 | 1.90 | 0.855 |
| Baseline high‐activation group (activation levels 3 and 4) | |||
| Patient activation level 3 in 2008 (baseline) | 1202 | 2.25 | |
| Patient activation level 3 in 2009 | 781 | 2.25 | 0.542 |
| Patient activation level 3 in 2010 | 740 | 2.20 | 0.570 |
| Patient activation level 4 in 2008 (baseline) | 1174 | 2.70 | |
| Patient activation level 4 in 2009 | 920 | 2.75 | 0.738 |
| Patient activation level 4 in 2010 | 941 | 2.60 | 0.850 |
When the data were restricted to patients who were present across all three surveys, results were unchanged. When the sample was restricted to participants who reported 10 or more PACIC items, the pattern of association between the PAM and the PACIC was also found to be almost identical to the one in the main analysis presented in Figure 1. Among the respondents who reported no change in activation level throughout all three survey years, the activation level was found to be positively associated with the PACIC in each year. The homogeneity of these findings across subanalyses lends further support to the finding from the main analysis.
Discussion
Surveys of patient experiences are useful in identifying elements within health‐care systems in need of change and are widely accepted and used in assessing the quality of delivered care.33, 34, 35 Our study used the PACIC instrument to measure, through patient experiences, what has traditionally been considered as the receipt of CCM‐congruent care. We demonstrate within a single health‐care system that changes in patient activation levels are associated with changes in the PACIC scores, with activated patients being more likely to report higher PACIC scores. This suggests that PACIC scores need to be interpreted in relation to patient activation.
The PACIC is designed to measure patient experiences with their chronic illness care.9, 10, 36 One reason for its positive association with the PAM could be that patients who moved to higher activation levels may actually be improving their interactions with care providers and thus enhancing the care they receive through better utilization of appropriate planned care, such as self‐management support (transactional hypothesis).37, 38, 39 This viewpoint gains support from studies that show patient activation to be positively associated with utilization of self‐management support services as well as life style behaviours, clinical outcomes, and quality of life,18, 40, 41, 42, 43, 44 or to be negatively associated with undesirable utilization of health services.21, 45 Despite this argument, it is also possible that more activated patients select providers who give better care and support (selection hypothesis). A previous study,46 however, concludes that there was evidence for the transactional hypothesis over the selection hypothesis.
Our findings highlight the need to consider both customer quality and technical quality in determining the overall quality of chronic care when patient perceptions are used. Thus, we recommend that PACIC be stratified according to patient activation levels. This recommendation for quality assessments concurs with the suggestion of Greene, et al.46 Also, assuming that the transactional hypothesis is true, then the relationship we found makes a strong case for considering patient activation/participation in improving the quality of chronic illness care. Thus, providers’ effort in improving patient activation during their provision of care can create a positive reinforcement cycle for improvement in the quality of care delivered. This kind of effort entails motivating, counselling and supporting patients to be more proactive in decision making, goal setting, problem‐solving, planned utilization of care and improving lifestyle and self‐care behaviours in managing their chronic condition. It goes beyond merely providing information or advice to patients and ticking the checklist of recommended care.
Utilization of motivational interviewing techniques in chronic illness care has been shown to be acceptable in patients with type 2 diabetes47 and feasible in a clinical setting.48 It was also found to improve patient activation and lifestyle behaviours in one study49 and self‐efficacy, self‐care, glycaemic control and quality of life in another.50 Furthermore, patients of general practitioners (GPs) trained in motivational interviewing reported receiving better care in terms of self‐care advice from their GPs and were found to be more ‘autonomous and motivated’ compared to patients of GPs without the training.51 The evidence regarding such interventions needs to be strengthened by prospective experimental studies, and further studies are needed to better understand the mechanisms of consultation varying by level of patient activation. Such studies can provide better insight into the transactional hypothesis and can lead to more targeted approaches to the issue of care improvement.
Our study is the first to look at the relationship between patient activation measured by the PAM and patient‐assessed quality of care measured by the PACIC using longitudinal data in patients with diabetes. The strength of our study is the 3 years of data with a large‐enough sample size for subgroup analyses. However, there are several data limitations. We had a poor response rate; although in previous surveys, it has been shown that bias due to low participation is not mandatorily present.52 Another issue common to surveys such as ours is attrition over time, but we demonstrate an unchanged pattern of associations in analyses including only patients who were present in all three surveys. In addition, sensitivity analyses for missing data did not impact on the analyses. We must, however, explicitly acknowledge that we are unable to draw any direct causal conclusion, although this was not the aim of this work.
We conclude that patient activation is an important factor to be considered when using patient experiences in quality of care assessments, such as that obtained with the PACIC. While this has been demonstrated among patients with type 2 diabetes, the findings reported here have implications for health‐care quality frameworks and quality performance indicators in chronic illness care. The association between improvement in patient activation and improvement in care experience implies that quality assessments that rely on the PACIC within or across health‐care facilities and indeed health‐care systems more broadly should be reported or compared within patient activation strata. Finally, we recommend that future studies look at the influence of patient activation on the relationship between patient‐assessed quality of care and clinical outcomes, and factors that link patient activation with access to and utilization of chronic care services to maximize patient outcomes.
Sources of funding
The Living With Diabetes Study was funded by Queensland Health. It was also supported in part by a Diabetes Australia Research Trust grant (20012000784) and the Australian Centre for Health Services Innovation (AusHSI).
Conflict of interest
All authors declare that there is no conflict of interest.
Acknowledgements
All authors contributed to the conceptualization of the manuscript. E.A. was the lead author in writing the manuscript. E.A. and S.A.R.D. analysed the data and wrote and edited the manuscript. M.D was responsible for the data acquisition and design of the study. S.A.R.D., J.R.C. and G.M.W. oversaw the data analysis. All co‐authors critically reviewed and edited the manuscript.
References
- 1. Stellefson M, Dipnarine K, Stopka C. The chronic care model and diabetes management in US primary care settings: a systematic review. Preventing Chronic Disease, 2013; 10: E26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Elissen AM, Steuten LM, Lemmens LC et al Meta‐analysis of the effectiveness of chronic care management for diabetes: investigating heterogeneity in outcomes. Journal of Evaluation in Clinical Practice, 2013; 19: 753–762. [DOI] [PubMed] [Google Scholar]
- 3. Wagner EH. Chronic disease management: what will it take to improve care for chronic illness? Effective Clinical Practice, 1998; 1: 2–4. [PubMed] [Google Scholar]
- 4. Glasgow RE, Wagner EH, Schaefer J, Mahoney LD, Reid RJ, Greene SM. Development and validation of the Patient Assessment of Chronic Illness Care (PACIC). Medical Care, 2005; 43: 436–444. [DOI] [PubMed] [Google Scholar]
- 5. Rick J, Rowe K, Hann M et al Psychometric properties of the patient assessment of chronic illness care measure: acceptability, reliability and validity in United Kingdom patients with long‐term conditions. BMC Health Services Research, 2012; 12: 293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gensichen J, Serras A, Paulitsch MA et al The Patient Assessment of Chronic Illness Care questionnaire: evaluation in patients with mental disorders in primary care. Community Mental Health Journal, 2011; 47: 447–453. [DOI] [PubMed] [Google Scholar]
- 7. Rosemann T, Laux G, Droesemeyer S, Gensichen J, Szecsenyi J. Evaluation of a culturally adapted German version of the Patient Assessment of Chronic Illness Care (PACIC 5A) questionnaire in a sample of osteoarthritis patients. Journal of Evaluation in Clinical Practice, 2007; 13: 806–813. [DOI] [PubMed] [Google Scholar]
- 8. Wensing M, van Lieshout J, Jung HP, Hermsen J, Rosemann T. The Patients Assessment Chronic Illness Care (PACIC) questionnaire in The Netherlands: a validation study in rural general practice. BMC Health Services Research, 2008; 8: 182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Glasgow RE, Whitesides H, Nelson CC, King DK. Use of the Patient Assessment of Chronic Illness Care (PACIC) with diabetic patients: relationship to patient characteristics, receipt of care, and self‐management. Diabetes Care, 2005; 28: 2655–2661. [DOI] [PubMed] [Google Scholar]
- 10. Schmittdiel J, Mosen DM, Glasgow RE, Hibbard J, Remmers C, Bellows J. Patient Assessment of Chronic Illness Care (PACIC) and improved patient‐centered outcomes for chronic conditions. Journal of General Internal Medicine, 2008; 23: 77–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Gugiu C, Coryn CL, Applegate B. Structure and measurement properties of the Patient Assessment of Chronic Illness Care instrument. Journal of Evaluation in Clinical Practice, 2010; 16: 509–516. [DOI] [PubMed] [Google Scholar]
- 12. Liu LJ, Li Y, Sha K, Wang Y, He X. Patient assessment of chronic illness care, glycemic control and the utilization of community health care among the patients with type 2 diabetes in Shanghai, China. PLoS One, 2013; 8: e73010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tabrizi JS. Quality of delivered care for people with type 2 diabetes: a new patient‐centred model. Journal of Research in Health Sciences, 2009; 9: 1–9. [PubMed] [Google Scholar]
- 14. Tabrizi JS, Wilson AJ, O'Rourke PK. Customer quality in health care. Patient Education and Counseling, 2009; 74: 130–131. [DOI] [PubMed] [Google Scholar]
- 15. Hibbard JH, Stockard J, Mahoney ER, Tusler M. Development of the Patient Activation Measure (PAM): conceptualizing and measuring activation in patients and consumers. Health Services Research, 2004; 39: 1005–1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Carryer J, Budge C, Hansen C, Gibbs K. Providing and receiving self‐management support for chronic illness: patients’ and health practitioners’ assessments. Journal of Primary Health Care, 2010; 2: 124–129. [PubMed] [Google Scholar]
- 17. Alegria M, Sribney W, Perez D, Laderman M, Keefe K. The role of patient activation on patient‐provider communication and quality of care for US and foreign born Latino patients. Journal of General Internal Medicine, 2009; 24(Suppl 3): 534–541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Remmers C, Hibbard J, Mosen DM, Wagenfield M, Hoye RE, Jones C. Is patient activation associated with future health outcomes and healthcare utilization among patients with diabetes? Journal of Ambulatory Care Management, 2009; 32: 320–327. [DOI] [PubMed] [Google Scholar]
- 19. Maeng DD, Martsolf GR, Scanlon DP, Christianson JB. Care coordination for the chronically ill: understanding the patient's perspective. Health Services Research, 2012; 47: 1960–1979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Tabrizi JS, Wilson AJ, O'Rourke PK. Customer quality and type 2 diabetes from the patients’ perspective: a cross‐sectional study. Journal of Research in Health Sciences, 2010; 10: 69–76. [PubMed] [Google Scholar]
- 21. Gerber LM, Barron Y, Mongoven J et al Activation among chronically ill older adults with complex medical needs: challenges to supporting effective self‐management. The Journal of Ambulatory Care Management, 2011; 34: 292–303. [DOI] [PubMed] [Google Scholar]
- 22. AIHW . Diabetes Prevalence in Australia: An Assessment of National Data Sources. Canberra: Australian Institute of Health and Welfare, 2009. [Google Scholar]
- 23. Donald M, Dower J, Ware R, Mukandi B, Parekh S, Bain C. Living with diabetes: rationale, study design and baseline characteristics for an Australian prospective cohort study. BMC Public Health, 2012; 12: 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. David M, Ware R, Donald M, Alati R. Assessing generalisability through the use of disease registers: findings from a diabetes cohort study. BMJ Open, 2011; 1: e000078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Aung E, Ostini R, Dower J et al Patient Assessment of Chronic Illness Care (PACIC) in Type 2 Diabetes: a longitudinal study. Evaluation and the Health Professions, 2014; 1–9 doi: 10.1177/0163278714556674. [DOI] [PubMed] [Google Scholar]
- 26. Hibbard JH, Mahoney ER, Stockard J, Tusler M. Development and testing of a short form of the patient activation measure. Health Services Research, 2005; 40: 1918–1930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Skolasky RL, Green AF, Scharfstein D, Boult C, Reider L, Wegener ST. Psychometric properties of the patient activation measure among multimorbid older adults. Health Services Research, 2011; 46: 457–478. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Skolasky RL, Mackenzie EJ, Riley LH 3rd, Wegener ST. Psychometric properties of the Patient Activation Measure among individuals presenting for elective lumbar spine surgery. Quality of Life Research, 2009; 18: 1357–1366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Stepleman L, Rutter MC, Hibbard J, Johns L, Wright D, Hughes M. Validation of the patient activation measure in a multiple sclerosis clinic sample and implications for care. Disability and Rehabilitation, 2010; 32: 1558–1567. [DOI] [PubMed] [Google Scholar]
- 30. Hung M, Carter M, Hayden C et al Psychometric assessment of the patient activation measure short form (PAM‐13) in rural settings. Quality of Life Research, 2013; 22: 521–529. [DOI] [PubMed] [Google Scholar]
- 31. Hibbard JH, Cunningham PJ. How engaged are consumers in their health and health care, and why does it matter? Research Brief, 2008; 1–9. [PubMed] [Google Scholar]
- 32. Campbell MJ, Gardner MJ. Calculating confidence intervals for some non‐parametric analyses. BMJ, 1988; 296: 1454–1456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Bruster S, Jarman B, Bosanquet N, Weston D, Erens R, Delbanco TL. National survey of hospital patients. BMJ (Clinical Research Ed), 1994; 309: 1542–1546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Elliott MN, Lehrman WG, Goldstein EH et al Hospital survey shows improvements in patient experience. Health Affairs, 2010; 29: 2061–2067. [DOI] [PubMed] [Google Scholar]
- 35. Cleary PD. The increasing importance of patient surveys. Now that sound methods exist, patient surveys can facilitate improvement. BMJ (Clinical Research Ed), 1999; 319: 720–721. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Horsburgh MP, Bycroft JJ, Mahony FM et al The feasibility of assessing the Flinders Program of patient self‐management in New Zealand primary care settings. Journal of Primary Health Care, 2010; 2: 294–302. [PubMed] [Google Scholar]
- 37. Adams RJ. Improving health outcomes with better patient understanding and education. Risk Management and Healthcare Policy, 2010; 3: 61–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Kane RL. What can improve chronic disease care? Journal of the American Geriatrics Society, 2009; 57: 2338–2345. [DOI] [PubMed] [Google Scholar]
- 39. Heisler M, Smith DM, Hayward RA, Krein SL, Kerr EA. How well do patients’ assessments of their diabetes self‐management correlate with actual glycemic control and receipt of recommended diabetes services? Diabetes Care, 2003; 26: 738–743. [DOI] [PubMed] [Google Scholar]
- 40. Rask KJ, Ziemer DC, Kohler SA, Hawley JN, Arinde FJ, Barnes CS. Patient activation is associated with healthy behaviors and ease in managing diabetes in an indigent population. Diabetes Educator, 2009; 35: 622–630. [DOI] [PubMed] [Google Scholar]
- 41. Munson GW, Wallston KA, Dittus RS, Speroff T, Roumie CL. Activation and perceived expectancies: correlations with health outcomes among veterans with inflammatory bowel disease. Journal of General Internal Medicine, 2009; 24: 809–815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Greene J, Hibbard JH. Why does patient activation matter? An examination of the relationships between patient activation and health‐related outcomes. Journal of General Internal Medicine, 2012; 27: 520–526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Hibbard JH, Mahoney ER, Stock R, Tusler M. Do increases in patient activation result in improved self‐management behaviors? Health Services Research, 2007; 42: 1443–1463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Mosen DM, Schmittdiel J, Hibbard J, Sobel D, Remmers C, Bellows J. Is patient activation associated with outcomes of care for adults with chronic conditions? Journal of Ambulatory Care Management, 2007; 30: 21–29. [DOI] [PubMed] [Google Scholar]
- 45. Begum N, Donald M, Ozolins IZ, Dower J. Hospital admissions, emergency department utilisation and patient activation for self‐management among people with diabetes. Diabetes Research and Clinical Practice, 2011; 93: 260–267. [DOI] [PubMed] [Google Scholar]
- 46. Greene J, Hibbard JH, Sacks R, Overton V. When seeing the same physician, highly activated patients have better care experiences than less activated patients. Health Affairs, 2013; 32: 1299–1305. [DOI] [PubMed] [Google Scholar]
- 47. Dellasega C, Anel‐Tiangco RM, Gabbay RA. How patients with type 2 diabetes mellitus respond to motivational interviewing. Diabetes Research and Clinical Practice, 2012; 95: 37–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Abramowitz SA, Flattery D, Franses K, Berry L. Linking a motivational interviewing curriculum to the chronic care model. Journal of General Internal Medicine, 2010; 25(Suppl 4): S620–S626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Linden A, Butterworth SW, Prochaska JO. Motivational interviewing‐based health coaching as a chronic care intervention. Journal of Evaluation in Clinical Practice, 2010; 16: 166–174. [DOI] [PubMed] [Google Scholar]
- 50. Chen SM, Creedy D, Lin HS, Wollin J. Effects of motivational interviewing intervention on self‐management, psychological and glycemic outcomes in type 2 diabetes: a randomized controlled trial. International Journal of Nursing Studies, 2012; 49: 637–644. [DOI] [PubMed] [Google Scholar]
- 51. Rubak S, Sandbaek A, Lauritzen T, Borch‐Johnsen K, Christensen B. General practitioners trained in motivational interviewing can positively affect the attitude to behaviour change in people with type 2 diabetes. One year follow‐up of an RCT, ADDITION Denmark. Scandinavian Journal of Primary Health Care, 2009; 27: 172–179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Nohr EA, Frydenberg M, Henriksen TB, Olsen J. Does low participation in cohort studies induce bias? Epidemiology, 2006; 17: 413–418. [DOI] [PubMed] [Google Scholar]