Savings from reducing low‐value general surgical interventions. BJS 2018; 105: 13-25.

Published: 8th November 2017

Authors: H. T. Malik, J. Marti, A. Darzi, E. Mossialos


Finding opportunities for improving efficiency is important, given the pressure on national health budgets. Identifying and reducing low‐value interventions that deliver little benefit is key. A systematic literature evaluation was done to identify low‐value interventions in general surgery, with further assessment of their cost.


A multiplatform method of identifying low value interventions was undertaken, including a broad literature search, a targeted database search, and opportunistic sampling. The results were then stratified by impact, assessing both frequency and cost.


Seventy‐one low‐value general surgical procedures were identified, of which five were of high frequency and high cost (highest impact), 22 were of high cost and low frequency, 23 were of low cost and high frequency, and 21 were of low cost and low frequency (lowest impact). Highest impact interventions included inguinal hernia repair in minimally symptomatic patients, inappropriate gastroscopy, interval cholecystectomy, CT to diagnose appendicitis and routine endoscopy in those who had CT‐confirmed diverticulitis. Their estimated cost was €153 383 953.


Low‐value services place a burden on health budgets. Stopping only five high‐volume, high‐cost general surgical procedures could save the National Health Service €153 million per annum.

Full text

Your comments

1 Comment

Jonathan Herron

1 year ago

It was with interest that we read Malik and colleagues review paper on savings from reducing low-value general surgical interventions. There are two important issues we wish to highlight.

Although there is no strong evidence either way for the use of facemasks to protect the patient, there is evidence they protect the surgeon. There are significant risks posed by mucous membrane exposure by HIV and hepatitis C, with the risk of blood splash being up to 45% in laparoscopic surgery (1). Should a surgeon choose not to wear a mask, they potentially commit an offence with the Health and Safety Executive (2008) stating that employees must wear a mask when blood splash is a possibility.

The current gold standard for appendicitis in older patients is a CT scan, particularly useful when there is a perforation (2) and other pathologies such as diverticulitis need to be excluded prior to surgical intervention. With the average medical negligence claim costing £165,836 only 26 of the 32,600 patients need make a successful average claim to be cost neutral with future potential for this figure to spiral (3).

Given the points highlighted in this letter, there remains doubt over the validity of the findings of the paper. Until these safety questions are unequivocally answered, good practice would be to continue to wear facemasks and to use CT scans as an adjunct for diagnosis of appendicitis. Other measures such as reducing pre-operative blood G&S samples (4) and selective histology for patients undergoing cholecystectomy (5) should be considered.

Herron JBT
Gilliam AD

1. Davies, C. G., Khan, M. N., Ghauri, A. S. K. & Ranaboldo, C. J. Blood and body fluid splashes during surgery – The need for eye protection and masks. Ann. R. Coll. Surg. Engl. 89, 770–772 (2007).
2. Carpenter, J. L. et al. Diagnostic Performance of US for Differentiating Perforated from Nonperforated Pediatric Appendicitis: A Prospective Cohort Study. Radiology 282, 835–841 (2017).
3. Fenn, P., Diacon, S., Gray, a, Hodges, R. & Rickman, N. Current cost of medical negligence in NHS hospitals: analysis of claims database. BMJ 320, 1567–71 (2000).
4. Thomson, P. M., Ross, J., Mukherjee, S. & Mohammadi, B. Are routine blood group and save samples needed for laparoscopic day case surgery? World J. Surg. 40, 1295–1298 (2016).
5. Emmett, C. D., Barrett, P., Gilliam, A. D. & Mitchell, A. I. Routine versus selective histological examination after cholecystectomy to exclude incidental gallbladder carcinoma. Ann. R. Coll. Surg. Engl. 97, 526–529 (2015).