Assessment of the quality of primary care for the elderly according to the Chronic Care Model

ABSTRACT Objective: to evaluate the quality of care provided to older people with diabetes mellitus and/or hypertension in the Primary Health Care (PHC) according to the Chronic Care Model (CCM) and identify associations with care outcomes. Method: cross-sectional study involving 105 older people with diabetes mellitus and/or hypertension. The Patient Assessment of Chronic Illness Care (PACIC) questionnaire was used to evaluate the quality of care. The total score was compared with care outcomes that included biochemical parameters, body mass index, pressure levels and quality of life. Data analysis was based on descriptive statistics and multiple logistic regression. Results: there was a predominance of females and a median age of 72 years. The median PACIC score was 1.55 (IQ 1.30-2.20). Among the PACIC dimensions, the “delivery system design/decision support” was the one that presented the best result. There was no statistical difference between the medians of the overall PACIC score and individual care outcomes. However, when the quality of life and health satisfaction were simultaneously evaluated, a statistical difference between the medians was observed. Conclusion: the low PACIC scores found indicate that chronic care according to the CCM in the PHC seems still to fall short of its assumptions.

In Brazil, the PACIC has been translated and adapted into Portuguese (6) , since the CCM has been incorporated by the Ministry of Health (15)  In light of the above, and assuming that older people with higher PACIC scores present better care outcomes, the objective of this study was to evaluate the quality of care provided to older people with diabetes mellitus and/ or hypertension in Primary Health Care, according to the Chronic Care Model, from the perspective of the elderly.
We sought specifically to identify associations between the overall PACIC score and the outcomes of the care provided in the sample studied.

Method
This cross-sectional study is part of the second phase of the population-based study "Aging and Renal Disease" (en-DoRen), whose overall objective in the Silva LB, Soares SM, Silva PAB, Santos JFG, Miranda LCV, Santos RM.
first phase was to estimate the prevalence of nondialytic chronic kidney disease in older people in one of the nine health districts of Belo Horizonte, Minas Gerais, Brazil (17) . The choice of this district was based on the fact that this had the highest absolute number of individuals aged 60 years or older (n = 44,801) at the moment of planning the first phase of the study. The en-DoRen study database identified 143 PHC users. Of this total, 118 elderly patients had a medical diagnosis of diabetes mellitus, hypertension or both, and were included in the current analysis. There was a loss of 13 people who did not respond to PACIC questionnaire due to death (n = 4), change of address and unsuccessful telephone contact attempt (n = 3), and lack of success to find the person at home after three unsuccessful attempts (n = 6). Therefore, the final sample of this study was composed of 105 older people.
It is noteworthy that these losses occurred due to the different chronology of approval from the Research Ethics Committee (REC) in the two phases of the en-DoRen study: before the REC approval to start the second phase, 54 elderly had already completed the first phase, and therefore, these elderly had to be approached a second time and invited to respond to the PACIC questionnaire.  almost never, 2) generally not, 3) sometimes, 4) most of the time, and 5) almost always (6) .
The mean overall PACIC score is obtained by the sum of the scores of each question, divided by the total number of questions (n = 20). In turn, the dimension scores represent the mean scores of the questions in each particular dimension. Higher scores indicate the perception, from the part of users, of greater involvement in self-care and greater support for the care of their chronic conditions (6) .
It should be emphasized that this questionnaire has been adapted and validated semantically and culturally by several groups interested in its use as a support tool for the diagnosis, adjustment, monitoring and evaluation of models of care to chronic conditions grounded in the Chronic Care Model, which has been tested in subjects with various chronic conditions, e.g. diabetes mellitus (11)(12)(13)(14)(23)(24) and cardiovascular diseases (23,25) .
Blood pressure and anthropometric data were measured within an interval of up to two weeks after the home visit for application of the questionnaire.
On this occasion, the collection of biological material (blood and urine) was scheduled for a maximum period of one week and guidelines were provided for the preparation of the test. The collection of biological material was performed in the morning by two members of the project, after a 12-hour fasting of the patient. The material was sent to a particular clinical laboratory for processing.
The dependent variables in this study were the overall score and the scores of each PACIC dimension.
The following variables were analyzed: sex (female, male); age in years; level of education in term of complete years of schooling (0-4 and 5 or more);  (30) . The biochemical variables collected were glycated hemoglobin (HbA1c), total cholesterol and fractions, triglycerides, fasting glucose, albumin/creatinine ratio (ACR), and serum creatinine.    (Figure 1).    q3. Asks about problems with the use of medication.
Delivery system design / decision support q4. Provides a written list of "things" that can improve health.
q5. Satisfaction with the organization of the treatment.
q6. Explains that whatever the user does to take care of himself influences the health problem.
Goal setting / tailoring q7. Asks about what the user wants to do to take care of the health problem.
q8. Helps to maintain healthy lifestyle habits.
q9. Provides a written care plan.
q10. Encourages the participation in specific groups.
Problem solving / contextual q12. The health team takes values, beliefs, and traditions into account.
q13. Helps to make the care plan.
q14. Helps to plan health care in difficult times.
q15. Asks about how the chronic condition affects life.
Follow-up / coordination q16. Gets in contact after consultation.
q17. Encourages the participation in community programs.
q18. Guidance on health care.
q19. Explanation of consultation with specialists/ help with treatment.
q20. Asks about visits to specialists.

Discussion
In this study, the overall PACIC score of 1.55 indicates that, in general, the congruence between the assessed care process and the CCM never occurred or occurred few times from the perspective of the participants. This result contrasts with other international studies that reached a total score higher than that presented in the current research, ranging from 2.33 to 4.19 (12)(13)(14)(23)(24)(25)32) . The only study at the national level, the one responsible for the translation of the questionnaire in the country, in Curitiba, reported a mean score of 2.86 (6) . Part of this difference can be explained by the fact that the cited studies included younger people with other morbidities.
Another relevant aspect is the possible influence of the presence of greater cultural homogeneity in other countries when compared to the Brazilian reality.
The low score found suggests weaknesses of Family Health teams in ensuring proactive, planned, coordinated and patient-centered care (6) . These flaws are reflected in possible difficulties in incorporating non-clinical aspects of chronic care into the practice, as for example, the implementation of assisted self-care (16) . This is one of the key elements of the CCM to ensure a high quality of care. The Pan-American Health Organization considers a fundamental and innovative strategy to assist people with chronic health problems.
Once the chronic condition is diagnosed, either diabetes mellitus or hypertension, the patient will need to deal with this condition in the daily life and, consequently, self-care will be a life-long task for the patient and his family (16) . To emphasize the importance of self-care, we stress that people with diabetes with the CCM's benchmarks than on technical quality, which is more easily evaluated (23) .
However, the low PACIC scores may also indicate that, even if these actions have been implemented In turn, when analyzing each individual PACIC item, it was found that only four questions obtained medians above 1.00 (q5, q8, q10, and q11). However, only q5 "Were you satisfied with the organization of your treatment" obtained a score above the average of the total possible score to be scored, with a predominance of the "almost always" response (36.2%), with the proviso that 21.9% of the elderly said they were "almost never" satisfied. This is because the discovery of a chronic condition requires that people change their daily life so as to be able to organize the care, from the development of skills to handle a range of activities whether or not predicted by medical knowledge, including adverse conditions for the control of the disease imposed by the socio-cultural context in which the patient is inserted, a situation called by some authors as "rupture of the biography of the individual".
However, over time (36) , patients may develop an attitude of "strategic lack of adherence", unconsciously and critically failing to comply with medical recommendations (36) . In this sense, some strategies Rev. Latino-Am. Enfermagem 2018;26:e2987.
they discover how to participate in the transformation of their world towards the integral health of the human being (37) .
As limitations of the study, we highlight the crosssectional nature of the study that makes it impossible to determine causal relationships of the outcome and variables of interest. There were also no national or international studies that included only elderly people using public health services for comparison purposes, and the bias of selective response cannot be excluded. As potentialities of the study, we highlight a population-based and randomized sample among the census sectors; the use of a questionnaire that has been translated and adapted to Portuguese, contributing to the accuracy of the answers given in the assessment of the quality of care; and the absence of missing cases in the PACIC questionnaire, which the literature indicates that can reach up to 32.7% (25,32) .

Conclusion
Poor quality of care provided for older people with diabetes mellitus and/or hypertension was found in the Primary Health Care according to the Chronic Care Model from the perspective of the elderly. This indicates that the reorganization of the care model oriented towards chronic care in the context of PHC seems still to fall short of its precepts, giving way to traditional biomedical models, from the perspective of the study participants.
It was not possible to confirm the hypothesis that older people with higher PACIC scores present better care outcomes, only in the case of older people who reported good or very good quality of life and who were simultaneously satisfied or very satisfied with health.
We recommend the expansion of strategies in loco that make it possible the diversification of prevention and management actions of health conditions that include the culture, values, and experiences of users.