Global Overview of Response Rates in Patient and Health Care Professional Surveys in Surgery: A Systematic Review

• Published in: Annals of surgery Vol. 275; no. 1; p. e75
• Main Authors: Meyer, Vincent Maurice, Benjamens, Stan, Moumni, Mostafa El, Lange, Johan F M, Pol, Robert A
• Format: Journal Article
• Language: English
• Published: United States 01.01.2022
• Subjects: Health Care Surveys - statistics & numerical data; Humans; Professional-Patient Relations; Surgical Procedures, Operative - ethics
• ISSN: 1528-1140
• DOI: 10.1097/SLA.0000000000004078

Identify key demographic factors and modes of follow-up in surgical survey response. Surveys are widely used in surgery to assess patient and procedural outcomes, but response rates vary widely which compromises study quality. Currently there is no consensus as to what the average response rate is and which factors are associated with higher response rates. The National Library of Medicine (MEDLINE/PubMed) was systematically searched from Januray 1, 2007 until February 1, 2020 using the following strategy: (((questionnaire) OR survey) AND "response rate") AND (surgery OR surgical). Original survey studies from surgical(-related) fields reporting on response rate were included. Through one-way analysis of variance we present mean response rate per survey mode over time, number of additional contacts, country of origin, and type of interviewee. The average response is 70% over 811 studies in patients and 53% over 1746 doctor surveys. In-person surveys yield an average 76% response rate, followed by postal (65%) and online (46% web-based vs 51% email) surveys. Patients respond significantly more often than doctors to surveys by mail (P < 0.001), email (P = 0.003), web-based surveys (P < 0.001) and mixed mode surveys (P = 0.006). Additional contacts significantly improve response rate in email (P = 0.26) and web-based (P = 0.041) surveys in doctors. A wide variation in response rates was identified between countries. Every survey is unique, but the main commonality between studies is response rate. Response rates appear to be highly dependent on type of survey, follow-up, geography, and interviewee type.