The decision this page changes. Someone trying to lose weight, or a clinician referring them, has to pick a format: a few check-ups with a GP, a monthly group, weekly sessions with a dietitian, a phone or web programme, a commercial programme. The single held source answers which formats move weight and by how much, and its answer is a contact-frequency gradient. Which diet goes inside the programme matters far less (-> Named Diet Programs Compared).

The source is the 2013 AHA/ACC/TOS adult obesity guideline (Jensen and colleagues, NHLBI Expert Panel). Its programme question, CQ4, rests on 51 trials (72 articles) rated good or fair after 74 poor-quality publications were excluded. Each CQ4 evidence statement (ES) carries a High / Moderate / Low grade (the guideline’s insufficient evidence statements, elsewhere, carry none). (Jensen et al., 2014)

The systematic search ran January 1998 to December 2009, with large RCTs (>100 per arm) published after 2009 added. (Jensen et al., 2014)

Everything here is weight, a surrogate. The CQ4 statements report kilograms lost and regained, not diabetes, events or function -> Surrogate Outcomes. What weight loss does to those outcomes is a separate question -> Does Weight Loss Reduce Cardiovascular Events, Lifestyle vs Metformin for Diabetes Prevention.

What “comprehensive” means: three components, all required

CQ4 ES1 (High) defines an effective high-intensity, on-site programme as diet plus physical activity plus behaviour therapy, all three included (Jensen et al., 2014):

  • Diet: an energy deficit of ≥500 kcal/d, often set as 1200-1500 kcal/d for women and 1500-1800 kcal/d for men.
  • Activity: aerobic activity such as brisk walking ≥150 min/wk; about 200-300 min/wk to keep weight off beyond a year.
  • Behaviour therapy: a structured programme with regular self-monitoring of food, activity and weight; for maintenance, weighing weekly or more often -> Self-Weighing for Weight Loss.

The grade describes what effective programmes contained. It does not separate how much each component contributes. (inferred from Jensen et al., 2014)

The intensity gradient — contact frequency is the dose

Intensity (on-site, comprehensive)Mean loss in the programme arm (vs usual care as each ES states it)GradeES
Low: less than monthly sessionsdoes not consistently produce weight loss vs usual careModerateES9b
Moderate: 1-2 sessions per month2-4 kg in 6-12 months; generally greater than usual careHighES9a
High: «frequent (ie, initially weekly)» on-site treatment (the ≥14-sessions-in-6-months figure is ES9c’s and rec 4b’s definition, not ES2a’s)up to 8 kg at 6 months, about 5-10% of starting weight; greater than usual careHighES2a
High, continued weekly-to-monthly for another 6 monthsup to 8 kg at 1 yearModerateES2b
Comparing the threehigh intensity gives greater net-of-control loss than low or moderate (the only kilogram ranking net of control; ES9b is a null against usual care)ModerateES9c

The kilogram cells are mean losses within the programme arms, not net-of-control differences, so the gap between the moderate and high rows is not itself a measured effect of the extra sessions; ES9c is the statement that compares net losses. (inferred from Jensen et al., 2014)

ES9a’s own wording on the usual-care contrast is hedged: «These losses generally are greater than those produced by usual care (ie, minimal-intervention control group).» (Jensen et al., 2014) ES2a’s population is adults «in whom weight loss is indicated and who wish to lose weight»: the high-intensity figures are for people who want to lose weight and have a reason to, not for everyone with a high BMI. (Jensen et al., 2014)

(Jensen et al., 2014) (ES2a-b); (Jensen et al., 2014) (ES9a-c)

The two key statements, verbatim: «ES9b (Low-Intensity Interventions): Low-intensity, on-site comprehensive lifestyle interventions, which provide less-than-monthly treatment sessions, do not consistently produce weight loss when compared with usual care.» and «ES9c (Effect of Intervention Intensity): When weight loss with each intervention intensity (ie, low, moderate, and high) is compared with usual care, high-intensity lifestyle interventions (≥14 sessions in 6 months) typically produce greater net-of-control weight losses than do low-to moderate-intensity interventions.» (Jensen et al., 2014)

The recommendations built on this are both Grade A (Strong): rec 4a, take part for ≥6 months; rec 4b, «Prescribe on-site, high-intensity (ie, ≥14 sessions in 6 mo) comprehensive weight loss interventions provided in individual or group sessions by a trained interventionist.» (Jensen et al., 2014)

Read the 14 sessions as a category edge, not a located knee. The three intensity bands are bins the panel compared, so the evidence shows more contact giving more loss across low, moderate and high. No statement locates a point above which extra sessions stop paying, and none tests anything above “high”. The guideline’s own gap list treats the ≥14-contact programme as «the standard for behavioral weight loss interventions» (Jensen et al., 2014), a reference point rather than a measured optimum. The studied range is the three bands; past high intensity there is no data here. (inferred from Jensen et al., 2014)

Who delivers it. A “trained interventionist” in the trials was mostly a health professional (dietitian, psychologist, exercise specialist, health counsellor) following a formal protocol, and in a few cases a trained lay person in a programme validated in published trials. (Jensen et al., 2014)

Primary care on its own: not shown to work

«ES5: In studies to date, low- to moderate-intensity lifestyle interventions for weight loss provided to overweight or obese adults by primary care practices alone have not been shown to be effective.» (High) (Jensen et al., 2014)

This is the most decision-relevant line on the page for the common case. ES5 concerns structured low- to moderate-intensity programmes run by practices, and the trials have not shown them to work; routine advice at ordinary visits is closer to the usual-care comparator than to these programmes. The guideline’s algorithm (Box 10) says «All patients for whom weight loss is recommended should be offered or referred for comprehensive lifestyle intervention (Box 11a and 11b).» (Jensen et al., 2014) Read together, the clinician’s job is to get the person into a structured programme, in the practice or by referral, rather than to counsel at routine visits. The ES says “not shown to be effective” for the trials reviewed; it is not a claim that no primary-care format could work. (inferred from Jensen et al., 2014)

Delivery mode — on site generally ahead of internet/email; telephone similar in one Low statement, and the guideline is split on remote delivery

ModeWeight resultGradeES / rec
Individual vs group sessions (high-intensity, on site)no substantial differenceLowES10
Telephone vs face-to-face (both with prepackaged meals or web support)similar, about 5 kg net of usual care at 6 and 24 monthsLowES4
Electronic (web) with personalised feedback from an interventionist, programmes developed in academic settingsup to 5 kg at 6-12 months (programme arm), more than minimal web or print adviceModerateES3
On site vs internet/emailon site generally greaterLowES11
Commercial, in person4.8-6.6 kg at 6 months (2 trials, conventional food); 6.6-10.1 kg at 12 months (2 trials, provided food)LowES6

(Jensen et al., 2014) (ES3); (Jensen et al., 2014) (ES4, ES6, ES10, ES11)

Both remote and commercial options are Grade B (Moderate) recommendations. Rec 4c: «Electronically delivered weight loss programs (including by telephone) that include personalized feedback from a trained interventionist† can be prescribed for weight loss but may result in smaller weight loss than face-to-face interventions.» Rec 4d allows commercial programmes «provided there is peer-reviewed published evidence of their safety and efficacy». (Jensen et al., 2014)

The commercial figures rest on two trials per row, all graded Low, and the 12-month row includes provided food, so they do not show that a commercial programme on its own matches a clinic programme. The mode comparisons are all Low: on site came out generally ahead of internet or email (ES11), while telephone matched face-to-face in the trials that tested it (ES4). Rec 4c hedges (remote delivery may give smaller losses). The web figure comes from programmes built in academic settings, which says little about whether a commercial app reaches it. (inferred from Jensen et al., 2014)

Added 2026-11-09. The algorithm text (Box 11b), unlike rec 4c, states the smaller loss without the hedge: «As found in CQ4, emerging evidence supports the efficacy, albeit with less weight loss, of electronically delivered interventions (eg, by Internet or telephone) that provide personalized feedback by a trained interventionist† and of some commercial programs that provide counseling (face-to-face or telephonic) with or without prepackaged meals.» (Box 11b; the next sentence makes referral to these options an expert-opinion recommendation, conditional on evidence of safety and efficacy) (Jensen et al., 2014) So the document reads three ways on remote delivery. ES4 (Low) finds telephone similar to face-to-face. Rec 4c, the graded recommendation (Grade B), says electronic delivery, telephone included, may give smaller losses. Box 11b, algorithm text, states the smaller loss without the hedge. The page cannot settle which reading the panel meant. On the held text, smaller loss is a possibility in the graded recommendation and a plain statement only in the algorithm, and telephone equal rests on one Low statement whose arms also had prepackaged meals or a web programme. (inferred from Jensen et al., 2014)

The trajectory: loss peaks at 6 months, then drifts back

The guideline is explicit that a programme’s peak is temporary. With continued contact after year one, «ES2c (Long-Term Weight Loss): Comprehensive lifestyle interventions that, after the first year, continue to provide bimonthly or more frequent intervention contacts, are associated with gradual weight regain of 1 to 2 kg/y (on average) from the weight loss achieved at 6 to 12 months. Long-term (>1 y) weight losses, however, remain larger than those associated with usual care.» (High) (Jensen et al., 2014) The diet question shows the same shape: 4-12 kg at 6 months, 4-10 kg at 1 year, 3-4 kg at 2 years (CQ3 ES3, High). (Jensen et al., 2014)

How many keep a meaningful loss: «ES8b: Of overweight/obese adults who participate in a high-intensity long-term comprehensive lifestyle intervention, 35% to 60% maintain a loss of ≥5% of initial body weight at ≥2 years’ follow-up (after randomization).» (Moderate) (Jensen et al., 2014)

Averages hide a wide spread. The algorithm text puts the expected average from a high-intensity programme at about 8 kg and says «The observed average weight loss of approximately 8 kg includes people who have variable weight loss (ie, some more and some less than average), so accurate prediction of individual weight loss is not possible.» (Jensen et al., 2014) That sentence sits in Box 11a, part of the treatment algorithm, and restates the evidence statements rather than adding a graded one.

The physiology behind the regain (appetite and energy-expenditure defence) is on Weight-Loss Maintenance and Metabolic Adaptation.

Maintenance: at least a year, and the optimum is unknown

Rec 4f (Grade A): people who have lost weight should take part long term (≥1 year) in a comprehensive maintenance programme. Rec 4g (Grade A) sets its content: contact monthly or more often, face to face or by phone, with 200-300 min/wk of activity, weighing weekly or more often and a reduced-calorie diet. (Jensen et al., 2014)

The evidence under it is Moderate. ES8a: a maintenance programme for up to 2.5 years after the initial loss reduces regain against minimal intervention, and «The optimal duration of weight loss maintenance programs has not been determined.» (Jensen et al., 2014)

So “≥1 year” is a floor inside a studied range of up to 2.5 years. It does not mean a year is enough; the trials did not test stopping points. (inferred from Jensen et al., 2014)

Very-low-calorie diets inside a programme

StatementResultGrade
ES7a: medically supervised VLCD (<800 kcal/d, meal replacements) in a high-intensity programme14.2-21.0 kg over 11-14 weeksHigh
ES7b: after the VLCD programme stopsregain of 3.1-3.7 kg over the next 21-38 weeksHigh
ES7c: moderate-intensity exercise (e.g. brisk walking 135-250 min/wk, aerobic or resistance) added to maintenance therapy after a VLCDdoes not reduce regain vs maintenance therapy aloneLow

(Jensen et al., 2014)

Rec 4e reserves VLCDs for «limited circumstances» under medical supervision. It is Grade A on the NHLBI scale, but the panel mapped it to ACC/AHA Class IIa rather than Class I. The footnote gives the reason in two sentences: «‡There is strong evidence that if a provider is going to use a very-low-calorie diet, it should be done with high levels of monitoring by experienced personnel; that does not mean that practitioners should prescribe very-low-calorie diets. Because of concern that an ACC/AHA Class I recommendation would be interpreted to mean that the patients should go on a very-low-calorie diet, it was the consensus of the Expert Panel that this maps more closely to an ACC/AHA Class IIa recommendation.» (Jensen et al., 2014) The bariatric rec 5a carries a parallel footnote (Grade A, mapped to IIa) on the same ground: «The ACC/AHA criterion for a Class I recommendation states that the treatment/ procedure should be performed/administered.» (Jensen et al., 2014) Revised 2026-11-09. Read with both footnotes, the IIa is a translation between two recommendation-strength scales. The NHLBI grade is a strength of recommendation, not an evidence level: Table 2 is headed NHLBI Grading of the Strength of Recommendations, and Grade A reads «There is high certainty based on evidence that the net benefit† is substantial.» (Jensen et al., 2014) The evidence level sits in the separate ACC/AHA Level of Evidence column, A for rec 4e, and it is unchanged too. Only the class moved: the NHLBI Grade A stands, and Class I would have meant should be performed, which the panel did not intend for a supervision-conditional recommendation. An earlier reading here called it a concern about how clinicians would read it rather than about the evidence. That is true, but it is a thin instance for Which Objective Moved This Recommendation: no non-health objective moved the recommendation, only its label on the second scale. (inferred from Jensen et al., 2014)

ES7c is about moderate exercise after a VLCD. A later trial of vigorous supervised exercise after an 8-week low-calorie diet is held on Weight-Loss Maintenance and Metabolic Adaptation; the two differ in intensity, comparator and timing, so they are not compared here.

Where drugs and surgery enter (comparator thresholds only)

The panel did not run an evidence question on drugs, because when the questions were set orlistat was the only approved weight-loss drug. (Jensen et al., 2014) Its drug recommendation is expert opinion: «The Expert Panel did not review comprehensive evidence for pharmacotherapy for weight loss. On the basis of expert opinion, the panelists recommend that for individuals with BMI ≥30 kg/m2 or BMI ≥27 kg/m2 with ≥1 obesity-associated comorbid condition(s) who are motivated to lose weight, pharmacotherapy can be considered as an adjunct to comprehensive lifestyle intervention to help achieve targeted weight loss and health goals.» (Jensen et al., 2014) The BMI cutpoint itself (≥30, or ≥27 with comorbidity) is not the panel’s: the algorithm footnotes attribute it to the FDA, «BMI cutpoint determined by the FDA and listed on the package inserts of FDA-approved obesity medications.» (Jensen et al., 2014) (Box 10 ‡; the same footnote is Figure 1 †, chunk 06). The expert opinion is the decision to adopt it. Bariatric referral (rec 5a, Grade A) is for BMI ≥40, or ≥35 with an obesity-related condition, after behavioural treatment has not reached the target; for BMI <35 there is no recommendation (5b, N). (Jensen et al., 2014)

The drug landscape has changed since: the class comparison that now sizes the drug alternative is on Comparing Obesity Drugs. This guideline’s drug position predates it and should not be read as a current comparator.

The held NICE review asks a different question

The one NICE document held, NG246 Evidence Review F, has a single review question: «What is the effectiveness and cost effectiveness of total or partial meal replacements, intermittent fasting, plant-based and low carbohydrate diets, in achieving and maintaining weight loss in adults living with overweight or obesity?» (National Institute for Health and Care Excellence, n.d.) It does not ask about programme intensity, contact frequency or delivery mode [searched: “contact frequency” 0 hits; “number of sessions” and “intensity” hit only trial descriptions, references and costing text, across the 10 Review F files]. NG246’s other evidence reviews are not held, so whether NICE answers this question elsewhere is unknown here. On this page’s question the wiki therefore holds one guidance family, and the parameter table that would compare two has no NICE row to fill -> Diets for Weight Loss - What NICE Recommends (where the diet-level comparison is). Gap (type-G): a current evidence review, with trials past 2010, on programme intensity and remote delivery. (inferred from Jensen et al., 2014; National Institute for Health and Care Excellence, n.d.)

Limits

  • Dated. Trial search to 2009 plus large later RCTs (>100 per arm). Remote and app delivery has grown since, so the remote-delivery rows are the most likely to have moved. Any divergence from newer guidance has lag as a live reading.
  • US setting. The guideline was written for US primary care. The held text does not say which trials sit behind ES5 or ES6, or where they ran, so how far the primary-care null and the commercial rows transfer to another health system is untested here.
  • Weight only. No CQ4 statement reports a patient-important outcome.
  • Bands, not a curve. The intensity effect is a comparison of three bands, each against usual care, not a dose-response fit.
  • Packages, not components. Each statement grades a whole programme; the share owed to contact, diet, activity or weighing is not separated.
  • Single source. One guideline family; the held NICE review does not ask this question (above).

(inferred from Jensen et al., 2014)

Every magnitude and grade on this page is the guideline’s summary of an evidence base the wiki does not hold. Added 2026-11-09. «This guideline is based on the Full Panel Report, which is provided as an online-only data supplement to the guideline.» and «The Full Panel Report contains background and additional material related to content, methodology, evidence synthesis, rationale, and references and is supported by the NHLBI Systematic Evidence Review, which can be found at http:// www.nhlbi.nih.gov/guidelines/obesity/ser/.» (Jensen et al., 2014) Neither the Full Panel Report nor the systematic evidence review is held, and this page cites no systematic review to check the ES figures against. The confidence: medium grades the guideline’s graded statements as reported, not an independent re-reading of the trials. (inferred from Jensen et al., 2014)

Decision relevance

  • Pick a format with at least monthly contact; the graded target is weekly at the start, ≥14 sessions in 6 months (rec 4b, Grade A). Less than monthly did not reliably beat usual care; 1-2 sessions a month gave 2-4 kg (High); weekly-start programmes gave up to about 8 kg at 6 months (High). These are programme-arm means; the net advantage of high intensity is ES9c (Moderate).
  • Routine GP advice alone is not a programme. Ask for referral to a structured one.
  • Telephone matched face-to-face in the trials that tested it (both arms also had prepackaged meals or a web programme); web with personalised feedback gave up to about 5 kg, generally less than on site. Both mode comparisons are Low. Rec 4c (Grade B) says electronic delivery, telephone included, may give smaller losses; only the algorithm text (Box 11b) states it without the hedge. Prefer on site where it can be kept up.
  • Plan for maintenance from the start. Keep monthly or more frequent contact for at least a year after the loss, with 200-300 min/wk of activity and weighing weekly or more (rec 4g). Expect 1-2 kg/y regain even with continued contact, and 35-60% to keep ≥5% off at 2 years (Moderate).
  • Group or one-to-one: whichever is easier to keep attending. The difference was not substantial.

(inferred from Jensen et al., 2014) — decision synthesis over the graded statements above.

References

Jensen, M. D., Ryan, D. H., Apovian, C. M., Ard, J. D., Comuzzie, A. G., Donato, K. A., Hu, F. B., Hubbard, V. S., Jakicic, J. M., Kushner, R. F., Loria, C. M., Millen, B. E., Nonas, C. A., Pi-Sunyer, F. X., Stevens, J., Stevens, V. J., Wadden, T. A., Wolfe, B. M., & Yanovski, S. Z. (2014). 2013 AHA/ACC/TOS Guideline for the Management of Overweight and Obesity in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and The Obesity Society. Circulation, 129(25_suppl_2). https://doi.org/10.1161/01.cir.0000437739.71477.ee
National Institute for Health and Care Excellence. (n.d.). Weight management: Evidence review F – effectiveness of different diets in achieving and maintaining weight loss. https://www.nice.org.uk/guidance/ng246/evidence