Multimorbidity in older adults: magnitude and challenges for the Brazilian health system

Multimorbidity frequency was high. At least 4 in every 5 major adults had ?2 morbidities
and 3 in every 5 had ?3 morbidities, thus confirming the importance of multimorbidity
as a frequent problem in older adults. The elevated number of dyads (n?=?22) with prevalence ?10 % and triads (n?=?35) with prevalence ?5 % highlights implications for the adequate management of
health problems in the same individual, with HBP being the problem most often associated
with other morbidities.

The percentage of multimorbidity found is consistent with the range of prevalence
encountered in two systematic reviews 25], 26] and recent studies 27], 28]. When considering only population-based studies, the frequency found in our analysis
was at least 10 percentage points higher with regard to the occurrence of ?2 morbidities
25]. The comparability of multimorbidity studies is hampered owing to methodological
differences, mainly related to the number of conditions included and the instruments
used to measure morbidity.

Achieving standardization is a challenge to the development of knowledge about multimorbidity.
References on this topic suggest that only chronic diseases should be included 1]. Despite the importance of acute conditions (e.g. influenza, tonsillitis and pneumonia),
which are more susceptible to seasonal variations, their inclusion tends to inflate
the occurrence of multimorbidity unnecessarily, thus complicating comparability 1]. Using at least 12 of the most prevalent morbidities appears to be advantageous because
they showed lower variability in multimorbidity frequency 25]. Similar to the decision taken in this study, a recent review suggested the inclusion
of certain geriatric syndromes in the construct of multimorbidity, such as urinary
incontinence and falls 1], considering their relevance for the quality of life and independence of older people
and for health care planning.

Therefore, taking the 12 most prevalent conditions in our data, multimorbidity frequency
was 78.4 % (95 % CI: 76.4; 80.5) for ?2 morbidities and 59.5 % (95 % CI: 57.0; 62.0)
for ?3 morbidities. After excluding urinary incontinence and falls from the morbidities
selected in this analysis, the frequencies were 77.1 % (95 % CI: 75.0; 79.3) for ?2
and 58.1 % (95 % CI: 55.6; 60.6) for ?3 morbidities. These findings are slightly lower
than those presented in results section. This reflects the low variability in the
occurrence of multimorbidity in the sample, regardless of the selected conditions,
and confirms the consistency of the prevalence found in this analysis.

The extrapolation of the data and its application to all the elderly living in the
city of Bagé intend to subsidize the health policies at SUS, providing an opportunity
for municipal health service management to plan actions for elderly people with multimorbidity.
This analysis takes into account the percentage of older adults with given characteristics
and this contributes to a more detailed evaluation to identify priority groups and
the magnitude of impact for future interventions, thus allowing the adequate planning
of actions aimed at these individuals. For example, the health care needs will be
relatively higher among residents in FHS catchment areas compared to residents in
traditional health service catchment areas in the city. Furthermore, the amount of
older adults living in FHS are bigger, increasing their relevance for health planning.
On the other hand, despite their low proportion, the management of multimorbidity
may become more complex in the bedridden elderly compared to those who are not bedridden,
eventually calling for more specialized care, and multidisciplinary teams.

Multimorbidity should not be seen as a major limitation of aging since its occurrence
is more a rule than an exception. The complications and interactions of multiple chronic
diseases represent a major challenge to the health services, because their impact
on the autonomy and independence of individuals 17], increasing the risk of disability and frailty 29], 30]. Complications are related to exacerbation of chronic health problems, for example,
uncontrolled high blood pressure that can lead to a stroke and increased risk of disability,
or the lack of control of blood glucose levels generating micro and macrovascular
problems closely related to the amputation of limbs.

The analysis by demographic, socioeconomic and health service type variables showed
the profile of the individuals most affected by multiple problems. Higher occurrence
among women may be attributed to survival bias since men tend to die earlier and those
who survive are usually the healthiest 4]. Another explanation is related to greater use of medical services by females 31] which was also observed in this study (data not shown) thus enabling more opportunities
for medical diagnosis of diseases. This results was similar to previous literature,
including populations of others age groups 32], 33].

The more elderly who mentioned having black or brown/yellow/indigenous skin color
had greater multimorbidity. This finding may be explained by the higher social and
economic vulnerability of these individuals in Brazil, highlighting social inequities
in health. The higher occurrence of multimorbidity among the more elderly is possibly
justified by a greater exposition to physiological stress and, then, to the occurrence
of chronic diseases.

The occurrence of multiple health problems was higher among older adults with less
schooling and lower income. This finding is similar to the large majority of studies
about multimorbidity 27], 33]–35], reinforcing the social determination of health and disease. Furthermore, it is worth
noting that Brazil is marked by inequalities in access to health services 15] and this could increases severity and complications.

The more elderly who used health services had greater multimorbidity. Reverse causality
is marked in these associations because elderly people with more health problems may
use more services or the use of services may have increased medical diagnosis. Nevertheless,
these associations may reflect the importance of health service utilization as a marker
of multiple chronic problems because, for example, almost all (95.1 %) the elderly
who used emergency services had ?2 health problems. Their relevance as a marker can
be an efficient way of quick screening elderly people with multimorbidity during assessments
by health professionals. Similarly, the bedridden elderly had more multimorbidity,
reflecting the greater vulnerability of these individuals.

The associations with health private plans and the PHC model reflect the focus for
actions directed towards management of multimorbidity. Elderly people without private
plans and living in FHS catchment areas had more multimorbidity. This confirms social
inequities since these elderly were poorer and less educated 21]. Whilst acknowledging that these actions may have been confused by socioeconomic
indicators, we believe that an adjusted analysis would not make sense for the purpose
of this article. Irrespective of confusion, individuals without a health plan and
living in FHS catchment areas have more diseases and greater social and economic vulnerability.
Thus, health actions related to the treatment and monitoring of chronic conditions
should prioritize these individuals.

The observed/expected ratios were statistically insignificant in four dyads (HBP/cognitive
impairment; spinal column disease/cognitive impairment; HBP/depression; and rheumatism/cognitive
impairment) and one triplet (HBP/spinal column disease/cognitive impairment) with
all having depression and cognitive impairment in the combinations. The measurement
of these two conditions was done by screening tests, which can increase the false
positive and reduce the specificity of combinations with causal relationship, as HBP
and cognitive impairment 36]. All the other 18 dyads with prevalence ?10 % and 34 triads with prevalence ?5 %
had a greater proportion than expected by chance. This reflects the occurrence of
morbidity clusters and a possible causal relationship between morbidities and/or risk
factors 37], 38].

However, the observed occurrence alone brings important information for clinical practice
and management of the health system and health services in Brazil. For example, approximately
one-fifth of the elderly have HBP and spinal column disease, thus indicating that
activities for the proper management of a health problem should take into account
all morbidities and not just one. For example, an elderly person with this pair of
diseases should be well instructed on how to engage in physical activities, since
although this is widely recognized as a good prognostic factor for HBP, it can also
aggravate back problems if undertaken without adequate guidance 39]. The simultaneous occurrence of HBP and cognitive impairment was observed in same
proportion as in the previous pair, thus highlighting the need for attention in the
approach used in the pharmacological treatment of these elderly people. The same rationale
is applicable to disease triads where morbidities and treatment interactions are more
important and increase the complexity of health care management.

Worldwide, health systems are still unprepared for the management of individuals with
multiple health problems and most guidelines are oriented towards a single disease
despite the occurrence of multimorbidity 12], 40]. The evidence presented here – added to the findings in the international literature
38], 41] – contributes to guiding the development and adaptation of Brazilian clinical guidelines.

In order to overcome the challenge of multimorbidity, the current fragmented health
care system for the elderly in Brazil should advance to a more comprehensive and multidimensional
care 42]. Goals to tackle chronic conditions have recently been established with the publication
of the strategic action plan to tackle NCD 14], the discussion on chronic care networks 13] and the consequent approval of the Ministerial Ordinance establishing the SUS Health
Care Network for People with Chronic Diseases 43]. However, these guidelines do not adequately include multimorbidity, mainly owing
to lack of information on the subject in the Brazil.

Promoting comprehensive care involving a considerable number of diseases, injuries,
conditions and complications is a complex task, which requires similarly complex answers.
The structure of a health system based on PHC is one of the leading measures to be
taken by countries to reduce inequities and improve health care efficiency 44]. In Brazil, these efforts largely depend on FHS universalization and effectiveness.

Some limitations of this study should be addressed. Multimorbidity operationalization
did not take into account the severity of the diseases, which could contribute to
the identification of priorities in the appropriate management of multiple health
problems. However, this approach would require greater detailing of disease severity
and for the purpose of this study the use of disease counts is considered more useful
than the use of scales/morbidities indices 45]. The other limitation is the absence of information about osteoporosis, thyroid disorders
and dyslipidemia and the lack of adequate information necessary to characterize some
morbidities. Although we have adequate measures for chronic morbidities through medical
diagnosis (e.g. hypertension and diabetes) and screening for cognitive impairment
and depression, we use proxies for other chronic morbidities, such as eyesight, hearing
and oral health problems.

Among its strong points, this is a population-based study with low probability of
selection bias in virtue of the low number of losses and refusals. Furthermore, the
sample characteristics are similar to the Bagé and Brazilian census of elderly population
collected in 2000 and 2010. These characteristics strength the internal and external
inferences about study, providing support to policy-makers of Bagé and similar Brazilian
cities in the actions related to multimorbidity. Moreover, the inclusion of a disease
set affecting different body systems (e.g. circulatory, visual and urinary systems)
have enabled a more complete approach to evaluate multimorbidity. Finally, reporting
the findings in accordance with recommendations in the literature may have contributed
to increasing comparability between studies.