Table 1.

Studies evaluated timing of referral to nephrology services with hospitalization and/or mortality as the outcomes

Paper number	Author(s)	Population	Study design and Quality Rating	Characteristics of Participants	Definition of late referral	Follow-up	Comments and results
1	Chan et al. 2007 [36]	22 studies (12,749 participants)	Meta-analysis QR = Fair	Studies with defined timing of referral and assessment of outcomes related to mortality or duration of hospitalization	1 to 12 months	NA	• 20 studies (total n = 12,018) that presented mortality data in early and late referred groups of subjects found that Late nephrology referral of CKD patients was associated with a significantly increased risk of death (relative risk [RR] 1.99; 95% CI [1.66, 2.39]; P < .0001) • 8 studies that assessed timing of referral and its impact on duration of hospital stay (total n = 3220) at the time of initiation of KRT found that late referred patients had a prolonged duration of hospitalization compared to the early referred patients by an average of 12 days
2	Stroupe et al. (2011) [48]	8,022	Retrospective study QR = Good	Kidney failure commencing KRT	< 3 months	1 year	• 37% received no pre-dialysis nephrology care, 24% received low, 16% moderate, and 23% high-intensity pre-dialysis nephrology. • During the year after dialysis initiation, patients in these groups spent an average of 52, 40, 31, and 27 days in the hospital (P < .001), respectively. • Greater intensity of pre-dialysis nephrology care was associated with lower costs even among patients whose first pre-dialysis nephrology visit was ≤ 3 months before dialysis initiation. • Patients with greater pre-dialysis nephrology care had lower mortality rates during the year after dialysis initiation (43%, 38%, 28%, and 25%, respectively, P < .001).
3	Kim et al. 2013 [49]	1,028	Prospective cohort study QR = Fair	Patients aged > 20 years starting dialysis for kidney failure	< 12 months	2 years	• Time from referral to dialysis was significantly longer in 599 ER patients than in 429 LR patients (62.3 ± 58.9 versus 2.9 ± 3.4 months, P < .001). • Emergency HD using a temporary vascular catheter was required in 485 (47.2%) out of all patients and in 262 (43.7%) of ER compared with 223 (52.0%) of LR (P = .009). • After 2 years of follow-up, the survival rate in ER was better than that in LR (HR 2.38, 95% CI [1.27, 4.45, P = .007]).
4	Smart et al. 2014 [50]	40 studies (63,887 participants)	Systematic review QR = Good	Studies of adult patients with CKD receiving dialysis with a known date on when they were initially referred to a specialist nephrology service	1 to 12 months	NA	• Compared to patients referred late, mortality was reduced across three, six, 12 and 60 months in those referred early • 4 studies reported mortality at three to four months data pooling showed mortality was reduced in patients referred early by 39% • One reported mortality at six months which was 42% lower in those referred early. • 16 studies reported 12-month mortality which was 35% lower in those referred early. • 3 studies reported five-year mortality data pooling showed mortality reduction of 33% in those referred early
5	Gillespie et al. 2015 [51]	443, 761	Retrospective cohort study QR = Fair	Patients starting dialysis or receiving a kidney transplant	< 12 months	1 year	• Overall, 33% of new KF patients had received no nephrology care, 14% had received < 6 months, 25% had received ≥ 6 but ≤ 12 months and 28% had received > 12 months of care prior to KF onset. • Any nephrology care prior to KF onset (versus no care) was associated with better health status and preparedness at the start of KF. • Longer pre-KF nephrology care was associated with lower first-year mortality (adjusted hazard ratio = 0.58 for > 12 months versus no care; 95% CI [0.57, 0.59]), higher albumin and haemoglobin, choice of peritoneal dialysis and native fistula and discussion of transplantation options.

ER: Early referred; LR: Late referred; KRT: Kidney replacement therapy; KF: Kidney failure; HD: Haemodialysis; QR: quality rating