The results of this study suggest that some predictors are related to users' self-reported disability due to spontaneous demand in primary health care. It is a relevant research for having the innovative perspective of studying the association of the type of care offered with the incapacity profile of users of the health system. In the general sample, users of family health units had less self-reported disability compared to those at the health center. It is understood that the model of organization of the family health type directs care centered on the subject, considering the degree of complexity required, and that, therefore, having a reference team for the users' demands, rather than professionals alternating in care, may have been a contributing factor to this finding.
In the study by Montoro et al. (2020)2, it was found that the predictors female gender, age ≥80 years, having ≥2 morbidities and having a self-perception of poor health status were routinely inserted as warning signs by the family health teams to track disabilities with the possibility of aggravation. In the study by Hustoft et al. (2019)²3, the longitudinality of care was a preponderant factor for individuals to report a lower level of disability in social participation and better self-perception of their health status. Thus, it is expected that effectively coordinated teams have an impact on the continuity of care and that patients experience better care on aspects of functioning when there are relational attitudes from the entire team, as is the case in the family health strategy24.
Regarding the prevalence of disability, 65.3% of family health users had some level of disability, compared to the prevalence of 80.9% in health centers. This finding is similar to that found by Montoro et al. (2020)2, who, in the same city of São Paulo, presented a prevalence of general disability of 56.4% in basic units affiliated to family health. However, it is a value well above the one found in the study by Naidoo et al. (2017)25, which found a prevalence of 38.9% in individuals aged 18 to 64 years with scores above 0 on the continuous scale of the WHODAS 2.0. These differences may be due to the location of the area and other context factors, as in the latter, the survey was carried out by clusters in households; so it is likely that the prevalence of disability is lower in a sample of the general population than in a sample that seeks care for a clinical care26,27. With regard to the severe level of disability, this study had a prevalence of 6.0% in the general sample, 7.2% in health centers and 2.3% in family health. In comparison with other studies, disability in a more general context was verified in the study by Salinas-Rodríguez (2020)1, which found 8.0% of severe diability in elderly people from low- and middle-income countries. In the context of samples with specific conditions, Karami et al. (2019)28 evaluated individuals with physical and intellectual disabilities and presented 28.9% of severe disability in their study. It is likely that those with more severe disabilities are less likely to participate in studies in a broader context29.
With regard to the prevalence of musculoskeletal pain, 59.0% reported having pain at the time of seeking care in primary health care units. This study presented an association between musculoskeletal pain and disability, so that answering “yes” to the presence of musculoskeletal pain increased 11.2 points in the continuous score of the WHODAS 2.0. The positive association between pain and self-reported disability has been discussed in several articles30,31,32. Although this relationship is not always observed in a proportional way, there may be functional improvement without monitoring the pain reduction and vice versa33,34.
In the multivariate evaluation of pain characteristics, only the intensity and number of pain sites remained significant in the final model. Despite the understanding that the characteristics of greater pain severity (worse intensity, more frequent, in different regions and more places) increase disability, when these are analyzed together, intensity becomes the main expression associated with the individual's disability. Users tend to refer more often to pain intensity than other characteristics when explaining how pain interferes with their daily life. Due to the lack of objective physical evidence of pain, the subject uses this characteristic to justify the existence of their pain to the other, in addition to adopting it as a mechanism to receive due attention and treatment, with a behavior related to medical and social expectations of pain related to the disease35. Silva et al. (2013)36 reported in their study that pain intensity, general and localized, had greater correlations with WHODAS 2.0 scores than other characteristics. For the authors, greater comprehensiveness of care, opposing fragmentation, can be attributed to the management of intensity, in the understanding that intervening in the reduction of global pain intensity is a better strategy than managing it in specific locations.
In addition to intensity, the number of pain sites was also relevant in this analysis. The dose-response effect with incapacity has also been verified in some studies12,20,36 indicating that multiple pain sites should be given greater attention in care to prevent greater severity of incapacity.
Per capita income was an important predictor of self-reported disability. This is in line with a previous study where WHODAS 2.0 scores were higher for lower-income participants37. Similar results were also found by Waterhouse et al. (2017)38, which revealed that the poorest income quintile was associated with severe disability and the number of chronic diseases reported. In general, individuals with generalized disability are more likely to occupy positions of low socioeconomic status, including not working, or working with low pay, having a lower educational level and lower family income39.
Age was also a factor associated with disability, so that having fewer years of life decreased the disability score when the multivariate regression model was analyzed, although this difference was not significant (p>0.05) in the bivariate analysis of the general sample These results diverged in most studies that assessed disability in the elderly29,37,38. However, a possible explanation is that in the primary health care setting, older individuals with more severe difficulties sought the units by spontaneous demand less than those individuals who were younger with the same degrees of perceived difficulty, which could suggest a worse access for elderly people with higher levels of disability. Another hypothesis is that these older users may have developed adaptation mechanisms that affect the perception of functioning, so that they may have faced their limitations in activities and participation as a normal consequence of aging. Anyway, more data would be needed to confirm these hypotheses.
In general, care and access to health must be guaranteed by the different types of care in primary health care and that health teams must adjust the most frequent demands with those with greater impact, dealing with phenomena of functioning, dependence, independence, illness and health, in an approach with the main guides of the biopsychosocial model of health.
STUDY IMPLICATIONS AND LIMITATIONS
The results suggest that the subjects' lower report of disability is indicative of better longitudinal care with the health service, so that the units that mostly have the family health reference team may provide greater surveillance of the conditions that most contribute to functional deterioration in their territory. In addition, understanding the characteristics of pain in this population can be useful to define assertive approaches to pain care that promote an improvement in disability and quality of life. Future studies can explore the relationship of assistential care as a causal factor for the functioning profile in a broader population.
This study has some limitations. First, it does not fully explore the disability profile based on the type of care in primary health care, as it was necessary for users to go to the collection units, therefore, it is possible that users with more severe disabilities were not interviewed. Another issue is the memory bias, so that the participants, in addition to reporting the intensity of pain at the time, were also asked to report it during the crisis, which did not always coincide with the pain the user had at the time of the interview. Finally, in this study, we did not verify the comorbidities of users in spontaneous demand, neither in order to account for them nor qualitatively classify them as possible predictors associated with disability. It is possible that these data could outline a better scenario of the profile of users who most seek care in primary health care, from the point of view of the health conditions that most interfere with self-report of disability.