Quantifying diagnostic intervals in multiple myeloma: a systematic review

Problem

Multiple myeloma is a common haematological cancer, with more than 4000 cases diagnosed each year in the UK. Due to the non-specific nature of the symptoms, patients with multiple myeloma can experience substantial delays in being referred to an appropriate specialist services. Such diagnostic delays often mean that patients present with more advanced disease with complications and start treatment late. It is widely believed that reducing the diagnostic delays will improve the relatively poor prognosis for people with multiple myeloma. The aim of this study is to review the published literature on diagnostic intervals in multiple myeloma across the whole diagnostic pathway and attempt to identify symptoms or combination of symptoms which are associated with the longest delays.

Approach

We searched MEDLINE and EMBASE databases up to February 2016 for full text articles containing data about diagnostic delays in symptomatic multiple myeloma. We excluded studies that included people aged less than 18 years and people with asymptomatic forms of the disease (Monoclonal Gammopathy of Undetermined Significance and Smouldering Myeloma). Our primary outcome was the clinical diagnostic interval (first presentation to final diagnosis) and secondary outcomes included the patient, primary care, secondary care and total intervals. All relevant descriptive statistics were extracted for each outcome. We report the median and interquartile ranges due to time not being a normally distributed variable. Meta-analysis was used to pool outcomes where possible and descriptive results were reported in other cases. Heterogeneity was examined in terms of symptoms and interval definitions. Risk of bias for the included papers was examined by using the Aarhus checklist.

Findings

750 articles were identified by our search of which nine provided data on diagnostic intervals. Study designs included cross sectional surveys, prospective patient studies and retrospective analysis of medical records. Five studies reported the clinical diagnostic interval, two studies reported the patient interval, one the primary care interval and one the total interval. No study reported the secondary care interval. Median diagnostic intervals ranged from 83 to 149 days across the five studies while the 25th and 75th percentiles ranged from 27 to 56 and 167 to 264 days respectively. Only 1 out 9 studies reported time intervals from specific symptoms. Further results will be available at the time of the conference.

Consequences

Our review suggests that up to a quarter of all multiple myeloma patients experience diagnostic intervals in excess of 5 months, but variability between studies is high. Our work shows that there is scope for meaningful reductions in the time to diagnosis but that more research is needed to identify the causes of the longer intervals.