Background
In primary care, patients with multiple chronic conditions are the rule rather than the exception [
1‐
3]. Compared to patients with single conditions, multimorbid patients are more likely to die prematurely, to be admitted to hospital and to have poorer quality of life [
4‐
6]. Although evidence exists that a structured, proactive and patient-centered approach helps to improve health outcomes, current delivery of care is often fragmented and event-driven [
7]. One widely accepted evidence-based framework for improving chronic care is the Chronic Care Model [
8,
9]. The CCM supports the provision of high-quality care and emphasizes the importance of continuity of care in a strong primary care sector. It aims to ensure care is planned, proactive and patient-centered, rather than reactive and focused on acute episodes, and it is designed to improve care in health systems at the community, organization, practice and patient levels. The model identifies key elements as essential to the provision of high-quality care to patients with chronic illnesses, i.e. self-management support, provision of clinical information systems, delivery system redesign, decision support, improved health care organization, and the use of community resources. For instance, the element “delivery system redesign” focuses on transforming a system that is essentially reactive into one that is proactive, thus ensuring the patient receives structured and planned care, as well as follow-up consultations, as part of a standard procedure [
10]. Interventions involving one or more elements of the CCM have shown beneficial effects on clinical outcomes and care processes [
11‐
13]. In Germany, little is known about the degree of implementation of elements of the CCM in routine primary care for multimorbid patients. The ‘Patient Assessment of Chronic Illness Care’ (PACIC) questionnaire can be used to assess patients’ perceptions in this respect [
14]. The PACIC has been used increasingly in primary care research in recent years [
15]. In an ongoing cluster-randomized controlled trial, for example, Van Lieshout et al. are using the PACIC to analyze the effectiveness of a tailored implementation program for patients with chronic heart failure in general practices [
16]. The PACIC questionnaire has been translated and validated in several languages, including German [
17] and Dutch [
18]. In Germany, the PACIC has also been used to evaluate disease management programs [
19] that have been implemented in primary care nationwide and are aimed at promoting evidence-based chronic care that contains similar core elements to those used in the CCM [
7]. When interpreting study findings based on the PACIC, it is important to understand what factors influence a patient’s assessment of care. Previous research indicates that a patient’s assessment of chronic illness care may depend, not only on the care received, but also on the patient him or herself. Cramm and Nieboer investigated patients with cardiovascular diseases and chronic obstructive pulmonary disease in the Netherlands and found that younger and less depressed patients report higher PACIC scores [
20]. In a large integrated health care delivery system, Glasgow et al., who developed the PACIC, found a slight correlation between PACIC values and age and gender of the treated patients, while education was not associated with them [
14]. In a sample of patients with osteoarthritis, on the other hand, Rosemann and colleagues found that female gender and age were weakly positively correlated with PACIC sum scores, while education was slightly negatively correlated with them. However, none of these correlations were statistically significant [
17]. Ludt et al., who analyzed a sample of patients with coronary heart disease receiving structured chronic care in a number of European countries, found that at the patient level, male gender, more frequent practice attendance, and fewer conditions, are associated with higher PACIC scores [
21].
The aim of this study was to describe how multimorbid older patients assess the routine chronic care they receive in primary care practices in Germany, and to simultaneously explore the extent to which practice- and patient-level factors determine patients’ assessments.
Discussion
This study shows that from the perspective of multimorbid patients elements of chronic care have not yet been fully implemented in primary care practices in Germany. While some key elements, such as delivery system redesign/decision support (e.g. ‘Satisfied that my care was well organized’) have been well implemented in routine care, other elements such as helping the patient to set specific goals and arranging follow-ups are less common. Patients’ assessment of care was not only associated with practice- but also with patient-level factors.
The mean PACIC score of 2.4 reported in this study is one of the lowest described to date. This probably reflects a “real life scenario” with respect to the care that multimorbid patients actually receive in Germany, as it is generally criticized for being fragmented and event-driven [
7,
9,
40]. Wensing et al. reported a similar mean score (2.3) in patients with COPD that had been treated in rural general practices in the Netherlands, but many other studies showed higher scores, with mean scores of 2.7 or 2.8 in samples of patients with cardiovascular disease [
21,
41], and scores of above 3 in patients with depression [
42]. The difference is probably attributable to the fact that most previous research has focused on patients that are enrolled in structured programs/DMPs (e.g. [
21]), or on patients receiving specific kinds of chronic care, such as case management (e.g. [
42]). Another possible explanation is that the patients in our sample presented greater disease severity due to their multiple chronic conditions. Delivering high-quality care to these patients may be difficult due to their complex needs [
43]. The finding that mean scores were highest for the subscale delivery system redesign/decision support and lowest for goal setting/tailoring is in line with recent findings in primary health care research, which have indicated a similar pattern in several other European countries [
20,
21], as well as in Canada [
44]. We found that only 12.9% of the total variance occurred at a practice level, and that most of the variance was due to differences between patients and random error. This would appear to be a general limitation in questionnaires on satisfaction or experience of care that we think deserves more attention in studies that use the PACIC. Salisbury et al. have shown that specific measures of a patient’s experience reveal differences between practices more clearly than do conventional measures of satisfaction [
45]. This is probably also true of the PACIC – in fact, the designers of the instrument have incorporated conventional questions relating to satisfaction (e.g. “satisfied that my care was well organized”) besides ratings involving specific events (e.g. “given a copy of my treatment plan”).
On the practice level, solo practices in this study were associated with higher PACIC scores than group or shared practices. This may indicate that smaller organizational units promote a stronger link between patients and providers. Lévesque et al., who analyzed the influence of different organizational models of primary health care on the PACIC, found that patients of solo practitioners and family medicine groups rated their care more highly than patients of specialist clinics [
44]. At the patient level, higher PACIC values were associated with having a DMP disease. This finding was to be expected, since, similar to CCMs, DMPs combine many elements of structured care. Szecsenyi et al. have shown that patients registered in the DMP for diabetes view their care as more structured than those who are not [
19]. It should be taken into account, however, that we had no information on whether an individual patient, who had a disease for which a DMP program was available, had actually enrolled in one. This may have led to an underestimate of the effect. However, since DMPs have been implemented successfully in Germany for more than 10 years now, and 6 million patients have enrolled in them, we can assume that practices nowadays provide more structured care to all eligible patients than they used to. Similar to Ludt et al. [
21], we found that higher PACIC values were also associated with more frequent GP contacts. In a systematic review of patient-physician relationships, Ridd et al. showed that regular contact to the practice appeared to be one of the main criteria determining how patients experience the care they receive, and that patients with long-term and complex problems prefer to consult a single doctor [
46]. Contrary to Cramm et al., [
20], we did not find an association between PACIC scores and either age, or depressive symptoms. This dissimilarity is probably attributable to differences in the analyzed samples: Our sample consisted of multimorbid patients with a mean age of 74 (SD 5) years, whereas Cramm et al. studied patients with cardiovascular disease and a mean age of 64 (SD 10) years, as well as patients with chronic obstructive pulmonary disease and a mean age of 66 (SD 11) years [
20]. The smaller standard deviation of age distribution in our sample is due to our inclusion criteria (age 65–84) and explains why our sample was more homogeneous in terms of age distribution than the sample of Cramm et al. This makes it more difficult to ascertain whether age could be a potential explanatory variable. One noteworthy and to our knowledge new finding in this study was that higher PACIC scores were associated with higher social support and higher autonomy in the patient’s past occupation. Muller et al. showed that patients with complex jobs that offer considerable autonomy are likely to have better health status [
47]. A study in Scotland found that people living in the most deprived areas are more likely to develop multimorbidity at a young age and to have greater mental health problems [
48]. The finding in this study that a patient’s characteristics influence how he or she experiences received chronic care raises the question whether this reflects different expectations, different ways of answering study questions or differences in the care provided to different types of patients within the same practices. This is an important subject and should be considered in further studies on quality of care [
45]. A recent review on strategies to improve health outcomes in multimorbid patients in primary care indicates that most studies fail to adequately consider the impact of socio-economic factors, and highlights the importance of considering the possible differential effects of interventions in different socio-economic groups [
49]. In this study, the percentage of practices with one or two physicians was 81%, which is typical for the setting. Such practices tend to work independently of each other and GPs often own the practice in which they work. In this setting, financial and personnel resources are often limited, and not comparable to academic or integrated settings in health maintenance organizations, where elements of the CCM are easier to implement. Over recent years, innovative ways of introducing elements of the CCM into German primary care practices have been investigated. One promising method is the involvement of specially trained healthcare assistants in the care of patients with chronic conditions [
3,
50,
51]. Healthcare assistants are employed in the majority of German primary care practices and are less qualified than nurses. Research has indicated that trained healthcare assistants can successfully work as case managers and perform interventions designed in accordance with the CCM, e.g. for patients with depression [
52] and patients with osteoarthritis [
53]. An ongoing study by Freund et al. will answer the question whether a complex, multifaceted intervention involving healthcare assistants as case managers reduces the likelihood of the (re-)hospitalization of multimorbid patients [
54]. Besides CCM-based interventions, other organizational or patient-oriented complex interventions were also shown to be effective in improving health outcomes among multimorbid patients in primary care [
49].
Strengths and limitations
The major strength of this study is that the prospective cohort study from which we took the baseline data is one of the largest studies on the characteristics and care of multimorbid patients in primary care in Germany. A comparison of non-participants with participants indicated that in terms of sex and the most common chronic conditions, the sample in this study was roughly representative of multimorbid older patients in primary care practices in Germany. Participants were, however, slightly younger than non-participants [
26]. The cross-sectional study design means we cannot draw causality assumptions, and this is a clear limitation when interpreting the results. A further limitation is that we could not calculate the overall mean PACIC score for 13.1% of all patients due to missing values. This proportion is relatively high and probably attributable to the fact that our sample of multimorbid older patients had problems understanding some of the questions. The problem of missing PACIC data has already been pointed out by Wensing et al., who reported that 22% to 35% of patients with diabetes or COPD treated in rural practices in the Netherlands did not provide answers to specific items in the PACIC [
18]. A comparison of the multilevel hierarchical modeling results (including cases with imputed data vs. cases with complete data) was reassuring in this regard, since the majority of findings remained very similar. Some uncertainty remains, however, in the interpretation of the association between PACIC and both autonomy of former occupation and weighted count of chronic conditions, since the results of the analyses with imputed data were different from those with non-imputed data in these two cases. Furthermore, the psychometric properties of the PACIC are the subject of ongoing discussion. While there is no doubt that the instrument itself is up to the task of assessing whether care is provided in accordance with the CCM [
15], major criticisms concern the ways in which factorial validity and internal consistency are evaluated [
15,
55]. Patients may also be reluctant to criticize their GP, especially if they have a good, long-term relationship with him or her [
46]. There is no single standardized way of measuring quality of care. In this study we have assessed the care received by measuring patients’ views. Since previous research has shown that patient satisfaction reports or evaluations of their healthcare experiences are not necessarily consistent with other measures of quality [
56], we do not assume that the inclusion of quality indicators would have strongly influenced our results.
Acknowledgements
The study was funded by the German Federal Ministry of Education and Research (grant numbers 01ET0729 and 01ET1006E).
This article was prepared on behalf of the MultiCare Cohort Study Group, which consists of Attila Altiner, Horst Bickel, Wolfgang Blank, Monika Bullinger, Hendrik van den Bussche, Anne Dahlhaus, Lena Ehreke, Michael Freitag, Angela Fuchs, Jochen Gensichen, Ferdinand Gerlach, Heike Hansen, Sven Heinrich, Susanne Höfels, Olaf von dem Knesebeck, Hans-Helmut König, Norbert Krause, Hanna Leicht, Melanie Luppa, Wolfgang Maier, Manfred Mayer, Christine Mellert, Karola Mergenthal, Anna Nützel, Thomas Paschke, Juliana J. Petersen, Jana Prokein, Steffi Riedel-Heller, Heinz-Peter Romberg, Ingmar Schäfer, Martin Scherer, Gerhard Schön, Susanne Steinmann, Sven Schulz, Karl Wegscheider, Klaus Weckbecker, Jochen Werle, Siegfried Weyerer, Birgitt Wiese, and Margrit Zieger. We gratefully acknowledge the support of Ferdinand M. Gerlach, Martin Beyer, and Andrea Siebenhofer for supporting the study in Frankurt. Many thanks also to Corina Güthlin for support in the preparation of this article and to Phillip Elliott for proof-reading the manuscript. We are grateful to the general practitioners who supplied the clinical information on their patients, namely Theodor Alfen, Martina Amm, Katrin Ascher, Philipp Ascher, Heinz-Michael Assmann, Hubertus Axthelm, Leonhard Badmann, Angelika Bakro, Horst Bauer, Veit-Harold Bauer, Sylvia Baumbach, Brigitte Behrend-Berdin, Rainer Bents, Werner Besier, Liv Betge, Arno Bewig, Hannes Blankenfeld, Harald Bohnau, Claudia Böhnke, Ulrike Börgerding, Gundula Bormann, Martin Braun, Inge Bürfent, Klaus Busch, Jürgen Claus, Peter Dick, Heide Dickenbrok, Wolfgang Dörr, Nadejda Dörrler-Naidenoff, Ralf Dumjahn, Norbert Eckhardt, Richard Ellersdorfer, Doris Fischer-Radizi, Martin Fleckenstein, Anna Frangoulis, Daniela Freise, Denise Fricke, Roland Friedmann, Nicola Fritz, Sabine Füllgraf-Horst, Angelika Gabriel-Müller, Rainer Gareis, Benno Gelshorn, Maria Göbel-Schlatholt, Manuela Godorr, Jutta Goertz, Cornelia Gold, Stefanie Grabs, Hartmut Grella, Peter Gülle, Elisabeth Gummersbach, Heinz Gürster, Eva Hager, Wolfgang-Christoph Hager, Henning Harder, Matthias Harms, Dagmar Harnisch, Marie-Luise von der Heide, Katharina Hein, Ludger Helm, Silvia Helm, Udo Hilsmann, Claus W. Hinrichs, Bernhard Hoff, Karl-Friedrich Holtz, Wolf-Dietrich Honig, Christian Hottas, Helmut Ilstadt, Detmar Jobst, Gunter Kässner, Volker Kielstein, Gabriele Kirsch, Thomas Kochems, Martina Koch-Preißer, Andreas Koeppel, Almut Körner, Gabriele Krause, Jens Krautheim, Nicolas Kreff, Daniela Kreuzer, Franz Kreuzer, Judith Künstler, Christiane Kunz, Doris Kurzeja-Hüsch, Felizitas Leitner, Holger Liebermann, Ina Lipp, Thomas Lipp, Bernd Löbbert, Guido Marx, Stefan Maydl, Manfred Mayer, Stefan-Wolfgang Meier, Jürgen Meissner, Anne Meister, Ruth Möhrke, Christian Mörchen, Andrea Moritz, Ute Mühlmann, Gabi Müller, Sabine Müller, Karl-Christian Münter, Helga Nowak, Erwin Ottahal, Christina Panzer, Thomas Paschke, Helmut Perleberg, Eberhard Prechtel, Hubertus Protz, Sandra Quantz, Eva-Maria Rappen-Cremer, Thomas Reckers, Elke Reichert, Birgitt Richter-Polynice, Franz Roegele, Joachim Rogalli, Heinz-Peter Romberg, Anette Rommel, Michael Rothe, Uwe Rumbach, Ariane Scheiding, Michael Schilp, Franz Schlensog, Ina Schmalbruch, Angela Schmid, Holger Schmidt, Lothar Schmittdiel, Matthias Schneider, Ulrich Schott, Gerhard Schulze, Heribert Schützendorf, Harald Siegmund, Gerd Specht, Karsten Sperling, Meingard Staude, Hans-Günter Stieglitz, Martin Strickfaden, Hans-Christian Taut, Johann Thaller, Uwe Thürmer, Ljudmila Titova, Michael Traub, Martin Tschoke, Maya Tügel, Christian Uhle, Renate Ulrich, Kristina Vogel, Florian Vorderwülbecke, Hella Voß, Christoph Weber, Klaus Weckbecker, Sebastian Weichert, Sabine Weidnitzer, Brigitte Weingärtner, Karl-Michael Werner, Hartmut Wetzel, Edgar Widmann, Alexander Winkler, Otto-Peter Witt, Martin Wolfrum, Rudolf Wolter, Armin Wunder, and Steffi Wünsch. We also thank Corinna Contenius, Cornelia Eichhorn, Sarah Floehr, Heidi Kubieziel, Rebekka Maier, Natascha Malukow, Christine Müller, Sandra Müller, Michaela Schwarzbach, Wibke Selbig, Astrid Steen, Miriam Steigerwald, and Meike Thiele for data collection, as well as Ulrike Barth, Elena Hoffmann, Friederike Isensee, Leyla Kalaz, Heidi Kubieziel, Helga Mayer, Karine Mnatsakanyan, Merima Ramic, Sandra Rauck, Nico Schneider, Jakob Schroeber, Susann Schumann, and Daniel Steigerwald for data entry.
Authors’ contributions
JJP, MAP and AD participated in the conception and data collection of the MultiCare study, performed the statistical analyses and drafted the manuscript. KM, JG, HH, SW, SRH, AF, WM, HB, HHK, BW, HVB and MS participated in the conception, design and data collection of the MultiCare study, and revised the manuscript critically for important intellectual content. All authors read and approved the final manuscript.