1 DRCR Retina Network Lapses in Care Among

Transcript:1 DRCR Retina Network Lapses in Care Among:
1 DRCR Retina Network Lapses in Care Among Patients Assigned to Ranibizumab for PDR: Protocol S Post Hoc Analysis Background DRCR Protocol S for PDR: PRP vs ranibizumab injections Mean VA with ranibizumab non-inferior to PRP over 5 yrs Less VF loss, ME, retinal detachments, vitrectomies over 5 yrs UK CLARITY: PRP vs aflibercept injections Mean VA with aflibercept superior to PRP (4 letters) at 1 yr Less ME, vitreous hemorrhage, adverse effects on VF, binocular VA Anti-VEGF therapy requires more patient visits than PRP Protocol S ranibizumab: median 43 visits, 19 PDR injections Protocol S PRP: median 21 visits, 50% no additional PDR treatment 2 Objective Describe patterns of visit completion in patients assigned to ranibizumab in Protocol S Identify factors associated with long lapses in care 3 Methods – Eligibility and Treatment Assignment Eligibility VA 20/320 or better No prior treatment for PDR Random assignment to PRP or ranibizumab 190 eyes assigned to ranibizumab 170 survived to 5 years => included in this analysis 4 Ranibizumab Treatment Protocol Initial treatment Injections Q4 weeks through 24 weeks (injections deferred at 16 and 20 weeks if no NV) At 24 weeks, injections Q4 weeks continue unless no NV or NV stable after 2 consecutive injections Restart a series of injections when NV worsened 5 Ranibizumab Follow-up Schedule Year 1 : Every 4 weeks Year 2 and later: Every 4 or 8 weeks if under treatment Every 16 weeks if under observation (extended to 20 weeks to accommodate annual study visits) 6 Lapse In Care by Scheduled Interval 7 Patient Profile 8 Time to First Lapse in Care and Dropping Out 9 85% 55% 29% Number of Episodes of Lapse in Care 10 1st long lapse resulted in dropout in 27% 27% dropped out later after returning from a long lapse Visual Acuity – Year 5 Completers 11 PDR-related Outcomes – 5 Year Completers 12 Timing of the PDR-related outcomes did not appear directly related to the timing of a lapse in care Baseline Risk Factors: Long Lapse in Care 13 C-statistic = 0.68 * P value from multivariable logistic regression model including all 3 factors Limitations Unknown whether dropouts sought care outside of the DRCR site. No information on the outcomes of the dropouts. Efforts to facilitate follow-up and financial incentives in Protocol S follow-up likely better than can be achieved in most clinical practices 14 Protocol S – Factors