A Quarter Portion, Not a Miracle

The strict dietary review asks a narrow question about if–then plans: when the cue–response component is separated from broader planning and tested against an active control, how much does adult fruit and vegetable consumption change? This chapter interprets that estimate in ordinary terms, then keeps its uncertainty, measurement limits, and unanswered clinical meaning attached to the number. The result is a useful calibration point for judging what “works” can honestly mean.

What the strict review actually tested

The label “implementation intention” covers more than one kind of intervention. Some studies ask people to link a recognizable situation to an actionable response. Others use broader action planning, such as deciding when, where, or how a behavior will occur, without clearly specifying a cue that can be perceived in the moment. If those forms are pooled without distinction, the resulting average may combine several mechanisms.

The review deliberately narrowed the category. An eligible plan needed a perceivable situation, a directly actionable and goal-directed response, and an explicit link between the two. The review excluded procedures that resembled conventional planning when one of those features was missing. This is not a claim that broader planning is useless. It is a way of asking a more specific causal question: what difference is associated with the strict cue–response component rather than with planning in general?

The comparison group mattered as well. The review included randomized controlled trials with active controls. The control participants received information comparable to the intervention group about the health benefits of increasing fruit and vegetable consumption and were asked to increase their intake, but they did not receive the strict if–then planning component under examination. That design makes the contrast more demanding than a comparison with no contact. A later difference cannot be explained as easily by the intervention group merely receiving a health message or being reminded that the behavior matters.

flowchart TD A["Candidate studies using planning labels"] --> B["Strict screen: perceivable cue, actionable response, explicit link"] A --> X["Incomplete or conventional planning excluded"] B --> C["10 articles, 12 comparisons, N = 2399"] C --> D["Randomized trials with active controls"] D --> E["Pooled mean difference: 0.29 portions per day"] E --> F["Standardized effect: 0.20, small"] F --> G["I-squared 32 percent; prediction can include a null future result"] G --> H["Retrospective self-report; practical value without established clinical meaning"]

The review therefore does not estimate the effect of a motivational package called “healthy eating.” It estimates a more isolated intervention component within a particular evidence base. That narrowness is a strength when the practical question is whether adding an if–then exercise is worthwhile. It is also a boundary: the result should not automatically be transferred to a different planning format, a different outcome, or a different population.

Ten articles contributed twelve intervention–control comparisons, with 2,399 adult participants. The intervention-to-outcome periods ranged from one week to 24 months. Those details tell us both how broad and how limited the evidence is. The number of participants is substantial enough to estimate an average across the included comparisons, and the follow-up range includes more than an immediate post-task measurement. But the studies are still a finite set of applied interventions, not a measurement of every context in which someone might try an if–then plan.

The narrow inclusion rule changes the meaning of the estimate in two directions.