Effectiveness of Integrated Care on Delaying Progression of stage 3-4 Chronic Kidney Disease in Rural Communities of Thailand (ESCORT study): a cluster randomized controlled trial

Chronic Kidney Disease (CKD) is a global public health problem, strongly associated with increased risk of death, cardiovascular events, and hospitalization [1‐3]. Despite several guidelines, its prevalence is still high in developing countries [4]. A recent community-based survey in Thailand revealed that the prevalence of stage 3–5 CKD was 8.8% [5]. The total number of pre-dialysis CKD patients in Thailand has been estimated at 4.1 million, which is impossible for the currently available 450 nephrologists to deal with. This nephrologist-to-patient ratio of 1:15,000 indicated that traditional physician-based approach is inadequate for ensuring equitable access to CKD care, even under the currently successful Universal Coverage scheme. With limited national resources, delaying the progression of CKD from early to advance stages is essential for efficient utilization of the limited number of nephrologists at minimizing healthcare cost. As patients residing in rural areas are less likely to reach well-qualified personnel when compared with city dwellers [6], it is mandatory to seek other appropriate forms of renal care to delay CKD progression.

Reorienting service delivery to focus on primary care and more efficient use of healthcare professionals other than nephrologists has been one of the promising solutions. Recent evidence suggested that a comprehensive intervention by nurse practitioners supervised by a qualified nephrologist could reduce the risk of cardiovascular events among diabetic patients as well as attenuate the decline of kidney function and improve the renal outcomes in CKD patients [7‐9].

Thai patients have enjoyed the unique primary health service delivery system in which health care officers at subdistrict level collaborated very well with village health volunteers (VHVs). The VHVs are villagers voluntarily recruited from their own villages to become key bridging points between subdistrict health and non-health systems. Approximately one million VHVs have covered more than 90% of all villages and therefore have become the backbone of community-based public health service in Thailand. This system has been so successful that Thailand was regarded as one of a few developing countries that have an efficient primary health care service [10‐13].

Should the primary health care officers and VHVs be trained to render proper CKD care, it would be interesting to examine whether their intimate relationship and commitment to their responsible village households will result in better outcomes than the conventional physician-based care. Evidence on the effectiveness of this type of comprehensive multidisciplinary intervention in developing countries has been lacking. One reason might be that this type of complicated intervention has been developed naturally out of initiatives of different stakeholders. Our attempt to clearly define the intervention allowed for further investigation of the outcomes.

This study was aimed to compare the effectiveness on delaying CKD progression between a conventional and an integrated CKD care (ICC) provided by a multidisciplinary care team (MDCT) of hospital staffs in conjunction with the a community CKD care network (CCN, subdistrict health care officers and VHVs).

Findings from this study are reported according to CONSORT guidelines [21]. Four hundred and forty-two participants met the enrollment criteria, and were included into the intervention (234 cases) and control (208 cases) groups (Fig. 1). The main reasons of non-participation (94 cases) were inability to attend the proposed visits, changing point of care to another hospital outside the study region and failure to contact the patients. The study was completed in July 2013 with 24-month median follow-up time. Sixteen patients (3.6%) were lost to follow-up, eight patients (1.8%) withdrew from the study. Most of baseline clinical and laboratory characteristics, including eGFR, of the two groups were comparable; however, the levels of HbA1c, 24-h urine sodium and normalized Protein Nitrogen Appearance (nPNA) of the control group were slightly higher than intervention group (Table 1).

This is the first, large scale, community-based randomized controlled trial evidence demonstrating that comprehensive community-based intervention by multidisciplinary care team in conjunction with a community care network teams of non-healthcare members can slow the rate of eGFR decline and is feasible and in CKD patients residing in resource-limited settings. Given similar medical care between the two groups including medication in particular ACEi/ARBs and educational materials, our study demonstrated that integrated CKD care significantly improved several clinical parameters with respect to serum bicarbonate levels, 24-h urine nPNA, 24-h urine Na, HbA1C, and serum triglyceride levels. Each parameter had been proven effectiveness on delaying CKD progression or lowering cardiovascular events by single-factorial intervention studies [22‐27]. In other words, we may imply that the key factors in delaying CKD progression of Integrated CKD Care were improvement of compliance with medication and dietary control. Consequently, this intervention seems to improve some non-conventional cardiovascular risk factors and may contribute to improvement of composite clinical endpoints. We expect the results of this study will set forth a new standard of community-based CKD care for Thailand and other resource-limited countries.

According to the results of several landmark studies, the key mechanisms that could improve clinical outcomes and delay rate of eGFR decline are optimization of traditional cardiovascular risk management. Gæde P, et al. (STENO2 study) demonstrated that a long-term (mean, 7.8 years), intensified multifactorial intervention in a specialized diabetes clinic, which targeted on blood pressure <130/80 mmHg, glycosylated hemoglobin <6.5%, triglycerides <150 mg/dL, and ACEi and Aspirin use, could significantly reduce risk of cardiovascular disease (HR 0.47; 95%CI 0.24–0.73) and nephropathy (HR 0.39; 95%CI 0.17–0.87) among patients with type 2 diabetes and microalbuminuria [7]. Peeters MJ, et al. (MASTERPLAN study) proposed that significantly improved rate of eGFR decline after 2 years of follow-up was mainly related to better BP control, increased use of ACEIs/ARBs, reduction of proteinuria, and possibly increased use of active Vitamin D [8]. On the contrary, our study revealed a different perspective of CKD treatments. At the baseline, mean systolic and diastolic blood pressure of the control and the intervention groups were not statistically different. Then during the study period, despite using the same treatment guidelines and medication, mean BP overtime of the control group was significantly lower than the intervention group. However, when looking at the mean BP overtime of the two groups, mean BP overtime of the two groups, both of them were within therapeutic range. Therefore, it would not be reasonable to interprete that lowering blood pressure could be a potential reason that contributes to the delaying of eGFR decline. Even though the changes of some parameters such as blood pressure, urine protein-creatinine ratio, and LDL-cholesterol levels did not concur with findings from prior multifactorial intervention studies, these parameters were within therapeutic targets according to standard guidelines.

Interestingly, it should be noted that baseline serum bicarbonate level of the two groups was not significantly different. In this study, we found that mean bicarbonate levels over time of the control group were significantly lower than the intervention group. Therefore, it might be implied that optimization of serum bicarbonate level could lead to the delaying of rate of eGFR decline whether this was due to the intervention or not. This finding is consistent with the finding of a randomized controlled trial which Goraya N, et al. demonstrated that treatment of metabolic acidosis in patients with stage 3 chronic kidney disease with fruits and vegetables or oral bicarbonate reduces urine angiotensinogen and preserves glomerular filtration rate [25, 26].

To our knowledge, even if the mean levels over time of HbA1c, triglycerides, PNA, and urine Na of the intervention group were lower, it is still unclear whether these factors affected the rate of eGFR decline as there were baseline differences among these variables. Based on an assumption that the longitudinal data analysis (GEE) might offset the baseline differences, these better-looking secondary outcomes might be implied that intervention could improve compliance with proper dietary protein and sodium intake and medication.

Importantly, increasing CKD awareness is another possible explanation why the integrated care could affect CKD treatment outcomes. In 2009, a population-based CKD screening study in Thailand, namely Screening and Early Evaluation of Kidney Disease (SEEK) study, revealed that only 1.9% of the 3459 participants were aware that they had CKD prior to the commencement of the screening program. As our intervention was ‘tailored’ to each patient, sustainable change of patient’s awareness and behaviors that affect renal and cardiovascular conditions could be anticipated [5, 28].

The collaborative effort between MDCT and CCN was the critical component of the intervention used in this study. While MDCT provided standard medical treatment using standard clinical guidelines and essential knowledge to CKD patients using live demonstration group counseling during each hospital visit, CCN played important roles in monitoring compliance to medication, diet control, blood pressure, and exercise behavior during a series of home visits at 6–8 weeks after each hospital encounter. We also saw the benefits of detailed discussion about some clinical parameters between the patient, his or her colleagues, and the CCN team members in a cozy environment. Clinical information was conveyed in a friendly way and questions could be addressed specifically to each patient in layperson terms, under supervision of healthcare providers. This approach is critical, especially when the majority of the CKD patients are not well educated. Healthcare providers also had a better understanding of living conditions and social dynamics of the community, which could affect the progression of CKD as well as other comorbidities.

This concerted approach might help explain possible mechanisms of the renal protective effect and the composite renal end points amongst patients in the intervention group. Evidence from clinical trials conducted in developed countries revealed satisfactory cardiovascular and renal outcomes from having multidisciplinary care in hospital settings [7‐9]. However, the intervention is not feasible and the findings therefore could not be generalized to developing countries where most of CKD patients lived in rural areas where specialized personnel is scarce and physician-to-patient contact time is limited [29]. While the best hospital-based CKD care is not equally available, we demonstrated that an optimal yet comprehensive set of harmonized efforts between healthcare providers and non-healthcare members is feasible and could yield similar, if not superior, clinical benefits. The implementation of the integrated CKD care model in developing countries is feasible. The use of community volunteers has been identified as one strategy to address the growing shortage of health workers, particularly in low-income countries. To our knowledge, the VHV scheme or Community Health Workers program has been established in more than 20 countries such as India, Brasil, Ethiopia, Kenya, etc. [30]. Furthermore, there is demonstrated evidence that a multifaceted intervention represents good value for money as it reduced costs but not quality of life for CKD patients (CanPREVENT study) [31].

With regard to the study which has been described above, there are some inevitable limitations. First, a cluster randomized control trial was chosen over a conventional double-blinded randomized control trial to account for the way service delivery was offered so it was possible for patients in the control group who reside in an overlapping area to be exposed to activities offered to the intervention group and vice versa. However, such phenomenon was believed to be minimal. Second, despite the randomization, there were several differences of baseline values including HbA1c, serum triglycerides, 24-h urine sodium and normalized Protein Nitrogen Appearance (nPNA), which could affect rate of eGFR decline. However, with the comparison of mean levels over time, the differences of baseline parameters may be offset. Lastly, distinguishing isolated effects of each component of our comprehensive set of interventions was difficult and therefore introduced some challenges to generalization of our findings to other settings that do not have the key CCN components.

In summary, the ESCORT study, a community-based, cluster randomized controlled study of 442 stage 3-4 CKD patients with a mean follow-up duration 2 years, shows that integrated CKD care may slow the rate of eGFR decline significantly and seems to improve the number of composite clinical events.