Hawthorne effect with transient behavioral and biochemical changes in a randomized controlled sleep extension trial of chronically short-sleeping obese adults: implications for the design and interpretation of clinical studies

Abstract

Objective: To evaluate the effects of study participation per se at the beginning of a sleep extension trial between screening, randomization, and the run-in visit.

Design: Subjects were screened, returned for randomization (Comparison vs. Intervention) after 81 days (median), and attended run-in visit 121 days later.

Setting: Outpatient.

Patients: Obese (N = 125; M/F, 30/95; Blacks/Whites/Other, N = 73/44/8), mean weight 107.6±19.7 kg, <6.5 h sleep/night.

Intervention: Non-pharmacological sleep extension.

Measurements: Sleep duration (diaries and actigraphy watch), sleep quality (Pittsburgh Sleep Quality Index), daily sleepiness (Epworth Sleepiness Scale), fasting glucose, insulin and lipids.

Results: Prior to any intervention, marked improvements occurred between screening and randomization. Sleep duration increased (diaries: 357.4 ±51.2 vs. 388.1±48.6 min/night; mean±SD; P<0.001 screening vs. randomization; actigraphy: 344.3 ±41.9 vs. 358.6±48.2 min/night; P<0.001) sleep quality improved (9.1±3.2 vs. 8.2±3.0 PSQI score; P<0.001), sleepiness tended to improve (8.9±4.6 vs. 8.3±4.5 ESS score; P = 0.06), insulin resistance decreased (0.327±0.038 vs. 0.351±0.045; Quicki index; P<0.001), and lipids improved, except for HDL-C. Abnormal fasting glucose (25% vs. 11%; P = 0.007), and metabolic syndrome (42% vs. 29%; P = 0.007) both decreased. In absence of intervention, the earlier metabolic improvements disappeared at the run-in visit.

Limitations: Relatively small sample size.

Conclusions: Improvements in biochemical and behavioral parameters between screening and randomization changed the "true" study baseline, thereby potentially affecting outcome. While regression to the mean and placebo effect were considered, these findings are most consistent with the "Hawthorne effect", according to which behavior measured in the setting of an experimental study changes in response to the attention received from study investigators. This is the first time that biochemical changes were documented with respect to the Hawthorne effect. The findings have implications for the design and conduct of clinical research.

Trial registration: ClinicalTrials.gov NCT00261898.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Study flow diagram.

Figure 2. Percent changes from screening in weight and waist circumference (WC), and in systolic and diastolic blood pressure (upper four panels, A–D), and the changes in sleep duration by diaries and by actigraphy watch (AW), and the changes in PSQI, and Epworth scores (lower four panels, E–H).

Error bars represent 1 SEM. Significant changes from screening: *P

Figure 3. Percent changes from screening in glucose parameters and in the prevalence of metabolic syndrome (upper four panels, A–D) and the changes in lipid levels (lower four panels, E–H).

Error bars represent 1 SEM. Significant changes from screening: *P