1. Introduction
Multimorbidity is defined as the co-occurrence of multiple (most commonly, two or more) long-term medical conditions in the same individual. It affects a substantial proportion of ageing societies [
1,
2]. The prevalence of multimorbidity varies from 20–30% in the whole population [
3,
4] to 55–98% in the elderly [
5].
Multimorbidity poses multiple challenges both to an individual and to health and social care systems [
1,
2]. For an individual, multimorbidity reduces the quality of life, physical functioning, life expectancy, and increases the risk of disability, psychological distress, as well as the risk of adverse drug reactions resulting from polypharmacy [
1,
4].
The current healthcare services are not adequately designed to meet the needs of patients with multimorbidity [
2]. Multimorbid patients often receive fragmented care, leading to inefficient, ineffective, and possibly harmful clinical interventions [
6]. For example, applying simultaneously several single disease-focused guidelines, which are based on data from randomized controlled studies in homogeneous patient groups, may result in overburdening the number of medicines, visits to healthcare institutions, rehabilitation procedures, and lifestyle modifications in multimorbid patients [
7]. Multimorbidity contributes to significantly higher healthcare utilization, e.g., it increases the expected use of primary and secondary healthcare services and the risk for unplanned potentially preventable hospitalization, emergency department admissions, and longer hospital stays [
8,
9].
Multimorbid patients’ healthcare in Lithuania follows global trends. From 2005 to 2019, the proportion of people aged 65 years and over increased from 15.8% to 19.8% [
10]. In a large cohort study, based on National Health Insurance Fund data for the period from January 2012 to June 2014, the prevalence of multimorbidity was 42% and 62% at the ages of ≥65 years and ≥85 years, respectively. Furthermore, more than 10% of the population already had at least two chronic conditions at the age of ≥45 years. Cardiovascular diseases (hypertension, ischemic heart disease, heart failure (HF), arrhythmias), diabetes, musculoskeletal diseases (osteoarthritis, back pain), dyslipidemia, stroke, and cancer were the most common chronic conditions accounting for multimorbidity [
11]. About 74% of healthcare resources in Lithuania are allocated to multimorbid patients who constitute about 20–30% of the total population [
10]. In the cohort study, multimorbidity was associated with an additional 258,761 inpatient days and increased the 30-day re-hospitalization rate by 61%. Outpatient visits in patients with multimorbidity were 2.1-fold more frequent compared with patients having a single disease. Patients with multimorbidity were 9.6-fold more likely to receive home visits by family doctors than patients with a single disease [
11].
Multimorbidity research includes not only objective outcomes such as mortality and disability but also subjective health outcomes such as quality of life, well-being and self-rated health [
12]. However, because high patient satisfaction is not necessarily equivalent to high quality of healthcare, a broader view is necessary to integrate patients’ subjective views and objective quality indicators into a comprehensive concept of good quality of healthcare [
13].
Various measures have been used to assess the quality of care in multimorbid patients. However, the reliance on measures oriented towards a single condition has been a major deficiency [
14]. Many measures have also been used to assess the quality of life and functional status in patients with multimorbidity in primary care. While these are particularly valuable for comparing the cost-effectiveness of interventions, they do not measure patient engagement, enablement and empowerment [
14].
A patient-centred approach increases patient satisfaction and counters the problems associated with fragmented healthcare, such as contradictory medical advice, overprescribing, over-hospitalization, and unresponsiveness. It requires a coordinated approach to healthcare organization and delivery [
15]. The Lithuanian healthcare system faces many issues that may particularly affect multimorbid patients: limited integration of public health and primary healthcare, inadequate coverage of family medicine services, insufficient expansion of outpatient services, insufficient coordination of responsibilities between the different healthcare levels, lack of payment model that incentivizes improvement of healthcare service quality, lack of tools to assess patients’ feedback, and insufficient patients involvement in the process of their treatment [
10]. Despite these challenges have been identified and recognized by health authorities, we lack information on how patients self-assess their needs and expectations regarding their morbidity and the healthcare problems they experience. Patient Assessment of Chronic Illness Care (PACIC) is a tool that has been increasingly used in several countries to measure how patients perceive the healthcare they receive.
This study assessed the multimorbid patients’ satisfaction with their healthcare quality and the factors that might affect their satisfaction.
4. Discussion
The mean PACIC+ questionnaire’s 5As summary score of 3.60 (0.93) in our population of multimorbid patients was relatively high compared with the scores in similar patients in other Western countries. The mean 5As summary scores established within the CHRODIS PLUS project before the implementation of ICMM were 2.91 (0.96) in Andalusia (Spain), 3.38 (0.54) in Aragon (Spain), and 3.17 (1.01) in Rome (Italy) [
17]. In American diabetic patients 62% of whom had two or more other chronic diseases, the mean 5As summary score was 3.2 (1.0) [
25]. In Greek diabetic patients 39% of whom had associated comorbidities, the mean 5As summary score was 3.1 [
26]. In German multimorbid patients with osteoarthritis, the mean 5As summary score was 2.52 (1.1) [
27]. Similarly, in German multimorbid diabetic patients, the mean 5As summary score was 2.78 (1.0) in patients not involved in diabetes management programs [
28]. In Western European patients with IHD having more than 3.3 comorbidities on average, the mean 5A summary score was 2.75 (95% CI 2.69–2.79) [
29].
The higher PACIC+ scores in our study can be explained, at least partially, by relatively frequent outpatient visits in Lithuania. The association between the frequency of visits and a patient’s satisfaction with their healthcare has been demonstrated previously [
30,
31,
32]. However, while the differences in 5As summary scores between Lithuania and Italy, Spain, or Greece (approximately 10 vs. 7, 7, or 3 visits per year, respectively) could be attributed to the higher number of yearly visits and hence the better opportunities for patients to ask questions, receive advice and support, this may not be a valid explanation for differences compared to Germany (approximately 10 visits per year) [
33]. Another reason for the higher PACIC+ score in our study population might be the proportion of participants from the outpatient departments affiliated with the university hospitals which had already applied the principles of integrated care for patients with multiple chronic conditions [
34] before the formal implementation of ICMM. Furthermore, our study population included a considerable proportion of participants with higher education (46.2%), whereas no more than 12% of participants had higher education in other study sites of the CHRODIS PLUS project [
17]. The potentially higher interest in the disease and its treatment options, better self-awareness, the ability to raise questions, and previous experience in setting goals and monitoring progress might have resulted in a proactive seeking of information and support.
Our study found decreasing patient satisfaction with their healthcare with advancing age. This trend was also observed in some [
19,
27,
35] but not all [
31,
32,
36,
37,
38] studies in patients with various chronic conditions, and in one study a positive correlation between PACIC results and age was observed [
18]. Although the age effect demonstrated in our study might be explained by decreasing patients’ cognitive abilities and determination to actively engage in their treatment, negative attitudes of medical personnel, reflecting the stereotypes regarding the older people prevalent in the region [
39] might have also played a role.
In contrast to some other studies where women rated better [
18,
35] or worse [
29] than men, we did not find an association between gender and PACIC+ scores. However, this finding is in line with the results of many other studies [
19,
25,
31,
32,
36,
37] and it may indicate that the impact of gender on satisfaction with healthcare is indeed negligible.
The results of our study suggest that the presence of depression has no impact on multimorbid patients’ satisfaction with their healthcare. In the PACIC validation study conducted on 255 patients with various chronic conditions, there were about 20% of patients with depression. The overall PACIC score in depressive patients and the scores for the five dimensions were not different from those in patients with other diseases [
18]. In a German study which specifically included primary care patients with major depressive disorder, the mean overall PACIC score was 3.25 (0.79) [
40], i.e., similar to the score range of 2.49 to 3.80 reported in patients with other chronic conditions [
29]. However, one study in diabetic patients (40% of whom also had arterial hypertension) found that depressive states assessed by Center for Epidemiologic Studies Depression Scale correlated with worse PACIC+ results [
26]. Thus, the influence of depression on patients’ satisfaction with their healthcare remains to be further clarified.
In our study, the PACIC+ scores, especially in the Assess, Advice and Agree subscales, were lower in patients reporting worse health-related quality of life. Likewise, higher quality of life (assessed by a 5-item scale: very good, good, neutral/reasonable, poor, very poor) was positively associated with PACIC scores in American patients with various chronic conditions and in Brazilian patients with diabetes [
41,
42]. Similarly, a study in Bosnia and Herzegovina found a higher overall PACIC score in hypertensive patients with self-perceived excellent health in comparison with those with self-perceived bad health [
31]. Self-perceived health in this study was measured by asking a question “What do you think about your health?” with the possibility to choose between three answers (excellent, good, or bad). There was a significant association between excellent health with higher scores on Delivery system design/Decision support, Goal setting/Tailoring, and Follow-up/Coordination subscales. Although PACIC+ subscales are not directly comparable with original PACIC dimensions (PACIC+ reflects the implementation of behavioural-counselling-based care, while PACIC reflects the implementation of original ICMM elements), the self-perception of healthcare in patients with lower quality of life may be particularly affected by the lower extent of information and support in problem solving provided to the patients (Decision support/Advise) and patients’ own engagement in their treatment (Goal setting/Agree).
Our patients with three or more chronic conditions had significantly lower PACIC+ 5As summary scores and individual subscales scores compared with patients having two chronic conditions. In the multicenter study in patients with IHD, assessment of healthcare quality was associated with the number of medical conditions, resulting in a 0.01 decrease in the PACIC score with each additional condition [
29]. An Australian study also found a significant association between the number of diseases and patients’ satisfaction with their healthcare, i.e., patients with diabetes and IHD/arterial hypertension had lower PACIC scores than those with diabetes only [
38]. A weak positive correlation between PACIC scores and the number of comorbidities was reported in the PACIC validation study [
18]. On the contrary, no association between patients’ satisfaction with their healthcare and the number of concomitant diseases was demonstrated in patients with type 2 diabetes [
25,
32,
36] or primary care patients in general [
43].
Although patients with more diseases may experience some benefits from more frequent visits and increased opportunities to receive information and help, they may also be more sensitive to healthcare fragmentation issues. Recent research has indicated that patients’ satisfaction is probably slightly higher in nurse-led primary care (moderate-certainty evidence), as well as their quality of life may be slightly higher (low-certainty evidence) [
44]. There are data supporting the opinion that depression, cognition treatment, and regular exercise can have a positive effect on patients’ satisfaction with their health [
45,
46].
Therefore, we advocate strengthening the role of a PHC team consisting of a general practitioner, a nurse or broad-scope nurse, a physiotherapist, and a lifestyle medicine specialist, and case management, developing the team’s competencies, and providing better testing possibilities. In addition, it is very important to ensure decision support systems for the PHC team members in the development of telemedicine services and remote general practitioner’s/specialist physician’s consultations.
We found that multimorbid patients’ satisfaction with healthcare depends on the presence of a specific comorbidity. In our study, hypertensive patients with IHD and HF showed lower scores on the Assess, Advise, and Agree subscales compared to patients without these diseases. Likewise, in the large multicentre study in Western European patients with IHD [
29], the Assess and Advise subscales, representing the level of gathering and providing information, rated highest, and the Arrange, representing organizational aspects of care, rated lowest. However, the score for the Agree subscale, representing the patient’s engagement, was highest in the above-mentioned study (2.98), while it was among the lowest (2.76) in our study. This may indicate that our patients with IHD may have received less endorsement to be actively involved in their own treatment. HF is a long-term debilitating disease that severely affects the quality of life by the need for hospitalizations, reduced ability to undertake daily living activities, and disturbed psychosocial well-being. The limitations in patients with HF may be even more expressed compared to those with other chronic conditions such as diabetes, cancer, or Alzheimer’s disease [
47]. This may have translated into lower satisfaction with healthcare in HF patients. To address these difficulties, The Health Ministry of Lithuania has issued a regulation establishing the role of a specialized HF nurse [
48]. When fully implemented, this program could significantly improve the healthcare of HF patients.
Patients with atrial fibrillation reported similar PACIC+ scores as patients without this comorbidity, except for the lower score on the Agree subscale. Studies evaluating the influence of atrial fibrillation on patients’ satisfaction with their healthcare using PACIC or PACIC+ questionnaires are lacking. European Patient Survey in Atrial Fibrillation (EUPS-AF), conducted prior to the approval and widespread uptake of direct thrombin and factor Xa inhibitors, found high patients’ satisfaction with their care: 85.5% of patients rated the quality of care on a five-point Likert scale as good, very good, or excellent [
49]. A recent study demonstrated mixed effects of atrial fibrillation treatment choice, with higher satisfaction in patients on direct oral anticoagulants in unadjusted analysis and higher satisfaction in patients on vitamin K antagonists in covariate-adjusted analysis [
50]. In Lithuania, most patients with atrial fibrillation use warfarin; therefore, they have to visit a PHC centre for coagulation tests once a month. Moreover, PHC institutions receive an incentive payment for coagulation assessments. More frequent visits might have resulted in better overall satisfaction with healthcare services, but the associated inconvenience might have negatively affected the Agree subscale. In addition, in Lithuania, direct oral anticoagulants are reimbursed only in patients with atrial fibrillation and two (in men) or three (in women) stroke risk factors [
51]. Limited opportunity to choose medications might have also contributed to the lower score on the Agree subscale.
Unlike patients with cardiovascular comorbidities, patients with diabetes had similar PACIC+ 5As summary scores as well as scores on the Assess, Advice, and Agree subscales compared with patients without diabetes. However, the mean scores on the Assist and Arrange subscales were significantly higher in diabetic patients (
p < 0.005). In the PACIC validation study, diabetic patients also achieved significantly higher overall PACIC, Goal setting/Tailoring, and Follow-up/Coordination scores [
18]. Similarly, the overall PACIC, the Goal setting/Tailoring, and the Follow-up/Coordination subscales resulted in higher scores in osteoarthritic patients with concomitant diabetes compared to patients with osteoarthritis only [
27]. During the validation of the Dutch PACIC questionnaire, diabetic patients reported higher satisfaction with structured chronic care in 14 out of the 20 PACIC items compared with patients with COPD. Higher scores were achieved in overall PACIC score and particularly in the Delivery system/Practice Design and Goal setting/Tailoring dimensions [
52]. It seems that diabetic patients are likely to receive more assistance with problem solving, treatment personalization, and arrangement of follow-up support. This may be associated with strict diabetes diagnostic and management protocols, availability of diabetes nurse consultations (up to four times per year or up to twenty-four times in those with diabetic foot), presence of educational programs, activity of patients’ organizations, and high overall attention to diabetic patients in Lithuania. The review of National Diabetes Plans in Europe conducted within the CHRODIS project emphasized the importance of political priority and adequate resource allocation for diabetes management. In this review, Lithuania’s policy was recognized as an example of the successful implementation of good practices from other countries and international guidelines, although admitting some limitations due to the lack of intersectoral collaboration [
53].
Patients with chronic diseases such as diabetes are continuously engaged with a PHC team, diabetes nurses and other specialists, including endocrinologists who consult patients with diabetes at least once a year. At a PHC level, patients with diabetes have to undergo glycated haemoglobin testing at least every 3 months. Moreover, regular visits and continuing care of these patients are conditioned not only by the competencies and specialization of PHC teams but also by financial incentive schemes (in Lithuania, a PHC institution receives incentive payment for glycated haemoglobin testing and diabetes specialist nurse consultations). Thus, in general, patients with diabetes attend outpatient clinics frequently and possibly due to that they are more likely to rate healthcare more favourably.
Our patients with musculoskeletal disorders showed low PACIC+ scores, especially on the Assess, Assist, and Arrange subscales. The German study in patients with osteoarthritis suggested that the lower Assist and Arrange scores may show the necessity to improve self-management support, including collaborative goal setting between doctors and patients regarding physical activity which is still underused in arthritis care [
27]. This may be also applicable to Lithuanian patients since the reimbursement of physical therapy is not sufficient to provide adequate support. Another reason for lower patient satisfaction may be less strict diagnostic and monitoring protocols for musculoskeletal diseases compared with those established for diabetes and cardiovascular diseases.
Patients with COPD or asthma were less satisfied with their healthcare compared with patients without these diseases as shown by lower Advise, Assist, and Arrange subscale scores. In the validation study of the Dutch PACIC version, patients with COPD rated worse than diabetic patients [
52]. In a Swiss study, low baseline PACIC+ scores significantly improved after the implementation of ICMM and educational programs [
54], suggesting that patients with respiratory system diseases may largely benefit from integrated care.
Possible reasons why patients with COPD, asthma, and HF were less satisfied with their healthcare might be the following: (1) PHC teams are not financially motivated to ensure regular care for patients with the above-mentioned chronic diseases and (2) PHC teams are not sufficiently specialized in continuing care of these diseases (e.g., there are no nurses who specialize in the provision of care for patients with HF and pulmonary diseases). In Lithuania, patients with HF are treated at tertiary healthcare centres providing specialized medical care. However, it is obvious that the accessibility of such services is limited, and these services should be provided at a PHC level. It is likely that an improvement in healthcare services for patients with pulmonary diseases and HF might lead to higher patient satisfaction in the future.
The major strength of this study is that it was the first study to provide information on how multimorbid patients perceive the quality of their healthcare. However, there were several limitations. Most patients were enrolled from the PHC centres associated with university hospitals; a large proportion of study patients were well educated. Further studies including more patients from different types of PHC centres are needed to extend these results to the general population of multimorbid patients in Lithuania. Furthermore, prospective studies assessing the effect of integrated care including general practitioners, nurses, case managers, mental health counsellors, and social workers could provide sound evidence for the directions in optimizing the healthcare of patients with multimorbidity.