The commentary by Killaspy ^{Reference Killaspy1} rather dismisses literature evidencing the value of the personal therapeutic relationship. It refers to a single publication which provides a qualitative theoretical classification of continuity issues by synthesising nine studies, most of which have no mental health component. The personal therapeutic relationship is the vehicle for delivering one of the most potent interventions in clinical medicine - the care (or ‘placebo’) effect. ^{Reference Moerman2}

Killaspy talks up the scientific basis for new developments, but the nature of randomised controlled trials is that they have significant exclusions which limit generalisability: the difference between efficacy and effectiveness. In particular, multimorbidity is common in the community and greatly diminishes the applicability both of a single trial and of guidelines which synthesise research findings. Killaspy appears not to respond to the issue that novel services developed by enthusiastic champions tend to lose efficacy when foisted on reluctant or inexperienced staff by government policy and/or managerial bureaucracy. She makes no reference to the poor implementation of proven research and the fact that government policies are not merely without evidence base but devoid of the mentality of scientific evidence. Scientists should be clear about generalisability, implementation and other caveats.

Further, the commentary does not answer the point that any change involving reduction in available beds will be associated with reduced bed usage. It claims that tariff-based healthcare will bring increased efficiency, whereas there is evidence that marketisation leads to financial inefficiencies and gaming the system, fragmentation of healthcare and blinkered specialism; ^{Reference Woolhandler and Himmelstein3} whereas what patients want is some continuity and someone to see the ‘big picture’.

The current multiplicity of teams inevitably increases interface issues which are often highlighted as causing problems in high-profile inquiries. It calls into question the claim Killaspy makes that ‘the service-line approach will reduce the need for many patients to move between services’.

I endorse the value of a therapeutic relationship with a single practitioner, particularly for long-term conditions (often multimorbid), and which often entails the other benefits noted by the commentary, including the efficiencies of personal knowledge standing astride balkanised interfaces. I do not wish to portray therapeutic relationships as a panacea free of side-effects - we know they are not always good and can even be damaging - but it is a recognised starting point with strong positive elements.

Of course, there are trade-offs between personal knowledge and other desiderata such as rapid access or specialist skills. We also know that re(dis)organisations have destructive elements and often overestimate the speed and magnitude of their benefits. ^{Reference Fulop, Protopsaltis, Hutchings, King, Allen and Normand4}

One conclusion might be that secondary care workers should abandon any intention to reap the benefits of continuity, and delegate this important role to our primary care colleagues. I personally consider that primary care, too, has its interfaces and discontinuities, and that mental healthcare for long-term conditions without long-term relationships would be sterile, soulless and counterproductive. As the National Health Service budget is being cut by 4% annually, the era of separate specialist teams may already be over.