Effectiveness of strategies to reduce low-value care: a systematic review of de-implementation studies




Poster session 2


Monday 17 September 2018 - 12:30 to 14:00

All authors in correct order:

Heus P1, van Dulmen S2, Weenink J3, Naaktgeboren C1, Verkerk E2, van der Laan M4, Kool T2, Hooft L1
1 Cochrane Netherlands, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, The Netherlands
2 Radboud University Medical Center, Radboud Institute for Health Sciences, IQ Healthcare, The Netherlands
3 Erasmus School of Health Policy and Management, Erasmus University Rotterdam, Rotterdam, The Netherlands
4 University Medical Center Groningen, Department Surgery, The Netherlands
Presenting author and contact person

Presenting author:

Pauline Heus

Contact person:

Abstract text
Background: Reducing low-value care is not straightforward and often hampered by barriers on different levels. Active strategies containing multiple interventions and addressing both healthcare providers and patients seem to have the highest potential for success. However, patients and clinicians are not the only stakeholders. There are also organizational and societal barriers to overcome. The impact of strategies also targeted at these barriers is unclear.
Objectives: To analyze which combinations of interventions, and the stakeholders they target (provider, patient, organisational context, and healthcare system), are most successful in reducing low-value care.
Methods: We searched MEDLINE, Embase and Rx for Change databases, and websites of healthcare organizations in April 2017, and checked references. English-language studies evaluating a strategy to reduce low-value care were eligible.
Results: We included 161 studies that aimed to reduce low-value care (de-implementation). Two-thirds (n = 108) were targeted at a single stakeholder, mainly a healthcare provider (n = 92; 85%). Of these, 70 (65%) strategies consisted of multiple interventions, usually containing an educational component and audit and feedback. Six (4%) strategies solely addressed patients. All distributed educational materials and one also used principles of shared management. Fifty-three (33%) strategies focused at multiple stakeholders: 34 (64%) at provider and context; 13 (25%) at provider and patient; four (8%) at patient and context; one (2%) at provider and healthcare system; and one (2%) at provider, patient and healthcare system. Overall, 134 (83%) studies succeeded in reducing the use of a healthcare service. Strategies targeted at multiple stakeholders were more often successful than strategies targeted at a single stakeholder (92% versus 79%, P value 0.028).
Conclusions: Successful de-implementation strategies were more often targeted at multiple stakeholders. This should be taken into account when designing de-implementation strategies. Just using multiple interventions is not always enough to reduce low-value care sustainably.
Patient or healthcare consumer involvement: There was no direct patient or healthcare consumer involvement in this systematic review, however, we do focus on the involvement of different stakeholders in de-implementation, including patients.

Relevance to patients and consumers: 

Healthcare with no or little additional clinical value for patients is called low-value care. This care should not be provided and strategies to reduce low-value care are being developed. It is not straightforward what the best strategy is, as barriers and facilitators associated with low-value care depend on the specific clinical context. It is therefore quite a challenge for doctors and policymakers to design the best strategies. Practical guidance is needed to compose and evaluate a tailor-made intervention for successful and sustainable reduction.