Mattress impact on lifestyle and back pain: a 25-year orthopedic literature review guideline

Abstract

Background

Sleep quality is influenced by several modifiable environmental factors, including the sleep surface. Low back pain, morning stiffness, shoulder discomfort, and muscle tightness on awakening may indicate inadequate bedding support. This systematic review evaluated whether mattress characteristics influence spinal alignment, sleep quality, and low back pain.

Methods

A systematic literature search and qualitative synthesis were conducted for studies published from 1999 to 2023. PubMed, Cochrane Library, Web of Science, Scopus, EMBASE, PsycINFO, ScienceDirect, and Google Scholar were searched using terms related to mattress ergonomics, pain, vertebral-column alignment, sleep, and quality of life. Eligible studies evaluated adult or young-adult populations and reported outcomes related to low back pain, spinal alignment, pressure distribution, sleep comfort, or sleep quality. Methodological quality and risk of bias were appraised using literature evaluation and grading of evidence (LEGEND) and Critical Appraisal Skills Programme (CASP) tools.

Results

A total of 47 studies were included. The evidence most consistently favored medium-firm mattresses over very firm or very soft surfaces. Medium-firm support was associated with better pressure distribution, improved sleep quality, reduced morning stiffness, and low back pain intensity. Zonal support, air-adjustable systems, customized designs, and thermal microclimate optimization showed potential benefits, but heterogeneity in materials, outcome measures, follow-up duration, and participant characteristics precluded meta-analysis.

Conclusions

Medium-firm mattresses represent the most defensible general recommendation for adults with nonspecific low back pain or morning stiffness. Mattress selection should also consider body habitus, age, sex-related anthropometry, preferred sleep position, and thermal comfort. Smart and adaptive mattress systems are promising but require larger, methodologically rigorous clinical trials before routine recommendation.

Level of evidence

Level III according to the Oxford 2011 Levels of Evidence.

Review registration

Not registered.

Trial registration

Not applicable.

Introduction

Sleep occupies approximately one-third of the human daily routine [1] and is essential for overall health, recovery, and daytime functioning [2]. Adequate, uninterrupted sleep supports metabolic regulation, neurocognitive recovery, immune function, hormonal release, and emotional balance [3, 4]. Although 8 hours of sleep is commonly recommended for adults, many individuals do not consistently achieve this duration because of occupational demands, shift work, lifestyle habits, stress, and environmental factors [5, 6]. Chronic sleep deficiency is associated with reduced quality of life and increased risk of metabolic, cardiovascular, musculoskeletal, and neurological disorders [7]. Therefore, the identification of modifiable factors that can improve sleep quality has relevant clinical and public-health implications.

The sleep surface is one of the most accessible and potentially modifiable components of the sleep environment. Mattress selection is often guided by subjective impressions of firmness, comfort, commercial claims, or the generic label of an “orthopedic mattress.” However, this label is not standardized and does not by itself guarantee biomechanical effectiveness or clinical benefit. Mattress characteristics may influence sleep quality through pressure distribution at the body–mattress interface, maintenance of neutral spinal alignment, support of the shoulder and pelvic girdle, facilitation of nocturnal movements, and regulation of the bed microclimate [8–13]. Inadequate bedding support may contribute to discomfort, lumbar pain, shoulder stiffness, muscle contracture, and perceived non-restorative sleep, particularly in individuals with low back pain, altered spinal mobility, or increased pressure sensitivity [14, 15].

The relationship between bedding support and low back pain is complex because it depends not only on mattress firmness but also on anthropometric characteristics, preferred sleeping position, age, sex-related body-shape differences, material properties, pillow–mattress interaction, and thermal comfort. Very firm surfaces may increase pressure peaks at the shoulders, pelvis, and sacral region, whereas excessively soft surfaces may allow the pelvis or trunk to sink, potentially disturbing neutral spinal alignment. Between these extremes, medium-firm support has been proposed as the most reasonable compromise between pressure relief and structural stability.

This systematic review was designed to answer the following PICO-framed question: in adults or young adults exposed to different mattress characteristics, compared with alternative sleep surfaces or personal bedding, which mattress features are associated with improved low back pain, spinal alignment, pressure distribution, and sleep quality? The objective was to synthesize the available orthopedic, biomechanical, ergonomic, and sleep-science evidence from 1999 to 2023 and provide clinically useful guidance for healthcare professionals advising patients with pain on awakening or sleep-related spinal discomfort.

Methods

This systematic review with qualitative synthesis included studies published from 1999 to 2023. The search covered publications available through 31 December 2023. Searches were performed in PubMed, Cochrane Library, Web of Science, Scopus, EMBASE, PsycINFO, ScienceDirect, and Google Scholar. Mendeley was used only as a reference-management tool and not as a bibliographic database. The search strategy was adapted to the indexing rules and syntax of each database. Medical Subject Headings (MeSH) were applied where appropriate, particularly for PubMed-based searches, and equivalent free-text terms were used in databases that do not support MeSH indexing.

The search terms included combinations of “mattress and ergonomics,” “mattress and pain,” “mattress and vertebral column,” “mattress and sleep,” and “mattress and quality of life.” Additional articles were identified by screening similar-article functions and reference lists of relevant publications. The search was designed to capture clinical, biomechanical, ergonomic, and sleep-science studies addressing the relationship between mattress characteristics and spinal or sleep outcomes.

Initial screening was guided by predefined eligibility criteria. Studies were eligible if they were written in English, published between 1999 and 2023, included adult or predominantly young-adult populations, and addressed low back pain, spinal alignment, sleep comfort, sleep quality, pressure distribution, or mattress-related biomechanical outcomes. Studies were excluded when the full text was unavailable, the publication date fell outside the defined period, the population was primarily pediatric or affected by conditions unrelated to the review focus, or the paper addressed topics not relevant to spinal alignment or low back pain. Papers focused exclusively on anti-pressure sore mattresses, ulcer prevention, humidity, ambient temperature, or isolated sleep-cycle measurements without spinal or musculoskeletal relevance were also excluded.

Methodological quality was assessed using the literature evaluation and grading of evidence (LEGEND) framework, while risk of bias was evaluated using Critical Appraisal Skills Programme (CASP) tools when applicable. Overall, methodological quality ranged from low to moderate. The most frequent limitations were small sample size, short follow-up, lack of blinding, heterogeneous mattress definitions, variable outcome measures, and the use of healthy volunteers rather than symptomatic populations.

The selection process followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 principles (Fig. 1).

Fig. 1.

PRISMA 2020 flow diagram summarizing the study selection process. The diagram reports the number of records identified, screened, assessed for eligibility, and included in the final qualitative synthesis, with reasons for exclusion at each stage.

After title and abstract screening, 479 potentially relevant papers were identified. Of these, 268 were excluded because they were duplicates or were considered irrelevant to the review question; including studies on unrelated populations such as rugby athletes, children, or pregnant women; and studies focused on unrelated outcomes such as mortality, allergies, rheumatologic conditions, or pressure ulcers. After abstract review, 103 additional papers were excluded. Of the 108 articles selected for full-text assessment, 61 were excluded because they did not directly address low back pain, spinal alignment, or clinically relevant sleep-surface characteristics. After applying the inclusion and exclusion criteria, 47 articles were included in the qualitative synthesis.

A standardized data-extraction table was created to collect author and year, study design, study aim, level of evidence, sample characteristics, methods, principal findings, and conclusions. These characteristics are presented in Table 1 [16–65]. The level of evidence of the selected studies was reviewed and harmonized between the manuscript and Table 1 as follows: one study was classified as level I evidence, seven as level II evidence, 35 as level III evidence, and four as level VI evidence. These categories were assigned according to the study design reported for each article and checked for consistency across the text and the evidence table.

Table 1.

Summary of included studies

Author (year)	Study design	Aim	Materials and methods	Conclusion
Kovacs et al. [21]	Randomized, blinded, controlled trial	To assess the effect of mattress firmness on chronic nonspecific low back pain	Level of evidence: I 313 adults with chronic low back pain lasting more than 3 months. Participants were allocated to medium-firm mattresses (n = 155) or firm mattresses (n = 158). Pain was assessed using a VAS; disability was assessed using the Roland–Morris questionnaire	Medium-firm mattresses were associated with greater improvement in chronic low back pain and disability than firm mattresses
McCall et al. [16]	Randomized controlled pilot trial	To compare sleep and discomfort on a traditional mattress and an anti-decubitus mattress over 14 days	Level of evidence: II 12 asymptomatic participants. Pain was rated using a VAS. Actigraphy and pressure mapping were used to evaluate sleep–wake patterns and contact-pressure distribution across seven pressure areas	No statistically significant difference was observed between the two mattresses, although the anti-decubitus mattress reduced peak pressure points
Bergholdt et al. [17]	Randomized single-blind clinical trial	To compare three mattress types over 30 days: waterbed (Akva), memory foam mattress (Tempur), and firm mattress (Futon Innovation)	Level of evidence: II 160 symptomatic participants. Pain and activities of daily living were assessed using the Danish COBRA questionnaire	Low back pain decreased more after sleeping on the waterbed or foam mattress than after sleeping on the firm futon mattress
Jacobson et al. [19]	Controlled trial	To evaluate the effect of mattress firmness on sleep comfort, spinal stiffness, and musculoskeletal pain	Level of evidence: III 22 symptomatic participants. Stiffness, sleep quality, upper-limb pain, and back pain were assessed using VAS scores	The selected bedding system was associated with reductions in shoulder pain, low back pain, and spinal stiffness
Jacobson et al. [20]	Controlled trial	To evaluate perceived back pain, stiffness, sleep discomfort, and comfort after introduction of a medium-firm bedding system	Level of evidence: III 59 adults. Pain and sleep discomfort were assessed using VAS scores; additional questionnaire items addressed irritation and perceived stress	A medium-firm mattress was associated with better sleep quality, pain relief, and comfort
Jacobson et al. [33]	Non-randomized controlled trial	To compare sleep quality and musculoskeletal symptoms on participants’ personal bedding and on a new mattress system	Level of evidence: III 59 healthy adults (29 men, 30 women). Sleep comfort was studied in two phases: 28 days on the personal mattress followed by 28 days on the new mattress. Stiffness, sleep quality, upper-limb pain, and spinal pain were assessed using VAS scores	The new mattress system promoted pain-free sleep comfort. The findings also supported timely replacement of worn bedding components
Jacobson et al. [34]	Controlled trial	To compare comfort, pain, and sleep quality after sleeping on a medium-firm mattress versus participants’ own mattresses	Level of evidence: III 59 healthy adults (29 men, 30 women). Pain and sleep quality were rated using VAS scores	Medium-firm bedding improved sleep quality and reduced low back pain
Jacobson et al. [35]	Controlled trial	To assess changes in pain and sleep after 3 weeks on participants’ own mattresses and 12 weeks on prescribed sleep surfaces	Level of evidence: III 27 symptomatic participants. Pain and sleep quality were rated using VAS scores	Prescribed sleep surfaces improved sleep comfort and reduced pain symptoms in participants with low back and/or shoulder pain
Monsein et al. [18]	A–B–A design	To compare participants’ own mattresses with an air-adjustable mattress topper in patients with low back pain	Level of evidence: III 90 participants with low back pain. Data were collected after the trial period on participants’ own mattresses and after 28 days of air-adjustable mattress use	The air-adjustable mattress reduced pain and improved sleep quality
Bader and Engdal [65]	Randomized controlled trial	To compare soft and firm mattresses and evaluate their effect on sleep quality	Level of evidence: II 10 asymptomatic participants. Sleep comfort was measured with polysomnography; pain and stress were assessed by questionnaire	No statistically significant differences were found between the mattresses
Lahm and Iaizzo [47]	Non-randomized controlled trial	To evaluate cervical and dorsal spinal alignment and physiological responses while resting on mattresses with adjustable air chambers	Level of evidence: III 29 healthy participants. Objective measurements at different mattress pressures included EMG activity, spinal alignment, sleep comfort, and cardiovascular parameters. Subjective opinions were collected using questionnaires	Changes in mattress pressure influenced vertebral alignment more than physiological parameters, supporting further study in symptomatic patients with back pain
Jin Park et al. [30]	Crossover trial	To assess compatibility between sleepers and mattress firmness using electromyography and subjective evaluation	Level of evidence: III 10 healthy asymptomatic participants. Mean age 31.7 years, mean height 170.9 cm, mean weight 70.6 kg. Participants used standardized pajamas. Spring and Tempur mattresses were compared. Muscle activity (RMS) was recorded by EMG from eight muscles, and eight subjective questions assessed hardness, comfort, and back support	The Tempur mattress showed higher RMS values, suggesting greater muscle activity. The spring mattress provided better perceived back support, whereas the Tempur mattress was preferred by participants seeking a greater cushioning effect
Raymann et al. [37]	Experimental study	To analyze the effect of skin-temperature manipulation during sleep	Level of evidence: VI 24 adults: 8 young asymptomatic participants, 8 older asymptomatic participants, and 8 participants with insomnia symptoms. A 4-day sleep protocol included home and laboratory nights. Skin temperature was controlled in a thermal suite during 6 h at night (31.7–34.5 °C)	Temperature influenced sleep comfort, particularly among older adults, who were more sensitive to thermal changes
Tonetti et al. [1]	Actigraphic study	To assess sleep quality over 2 weeks on spring mattresses and Myform-system mattresses	Level of evidence: VI 28 asymptomatic participants (14 men, 14 women). Objective measures included sleep parameters, insomnia duration, and muscle activity. Subjective measures included the Hassles Scale and the Mini Sleep Questionnaire	The Myform system tended to improve sleep parameters, but the differences were not statistically significant
Cheng et al. [46]	Crossover trial	To analyze mattress comfort using body-pressure distribution, spinal curvature, and subjective comfort scores	Level of evidence: III 10 healthy adults (5 women, 5 men), aged 20-43 years. Two same-size mattresses were compared: a traditional mattress (OM) and an adaptive double mattress (AM) with an adjustable right side. Spinal alignment in standing and left lateral lying positions was evaluated using the Qualisys system. Participants rated mattress characteristics and body-region comfort on a 1–7 scale	The adaptive mattress provided better support and lower pressure than the traditional mattress, especially at the buttocks and back. Both systems had limitations in upper-body support and spinal-curvature control; the adaptive mattress was more comfortable at the waist
Verhaert et al. [53]	Non-randomized controlled trial	To analyze sleep changes while using an ergonomic mattress system	Level of evidence: III 17 asymptomatic participants. Objective measurements included polysomnography and video recording. Subjective assessments included VAS, Karolinska Sleepiness Scale, Cox Stress/Arousal scale, Profile of Mood States fatigue scale, and Adjective Check List	Comparison of two systems alone was insufficient to define the effect of the bed on sleep. Lateral sleeping was associated with a shorter REM period, and the relaxed system negatively affected sleep in this sample
Park et al. [36]	Multicenter controlled trial	To assess relationships among mattress characteristics, anthropometry, body-pressure distribution, and spinal curvature	Level of evidence: III 18 asymptomatic adults (9 men, 9 women). Mattress pads used six materials (three cotton types, felt, sponge, and elastic cotton) and springs with 3-, 4-, or 5-pitch configurations. Three-dimensional reconstruction of the vertebral column and pressure sensors assessed peak pressures at the hips and shoulders. Six subjective 7-point questionnaires were administered	The best mattress maintained spinal curvature in lying positions closer to the standing posture. Comfort improved as firmness increased within the tested range
Shen et al. [57]	Randomized controlled trial	To analyze 18 spring-mattress configurations using five spring cores and 14 layer combinations	Level of evidence: II 8 healthy participants. Questionnaires assessed fatigue, pain, and well-being before and after sleep. Objective measurements included polysomnography and actigraphy	Mattress structure influenced sleep quality and vertebral alignment
Lee and Park [63]	Randomized controlled trial	To evaluate the influence of mattress type on sleep quality and skin temperature using polysomnography	Level of evidence: II 16 asymptomatic participants aged 20–30 years. Assessments included personal evaluation, polysomnographic measurements, and skin-temperature monitoring	A mattress designed to reduce unnecessary movements, support the spine, and maintain comfortable body temperature may improve restorative sleep
DeVocht et al. [15]	Controlled trial	To compare four high-end commercial mattresses from different manufacturers using biomechanical measures	Level of evidence: III 18 healthy men. Body-pressure distribution in lying position and spinal alignment in lateral position were measured	Higher pressure was recorded at the pelvic girdle than at the chest. Mattress A showed the lowest pressure distribution, whereas mattress D showed higher peak pressure and lower spinal distortion
Yu-Chi et al. [25]	Randomized crossover trial	To estimate the influence of topper thickness and hardness using objective physiological and biomechanical parameters	Level of evidence: III 40 healthy men divided into four BMI groups; mean age 22.2 years (range 18–28 years). Participants without back pain or relevant disorders used standardized pajamas in the supine position. Twelve toppers combined three hardness levels (soft, medium, hard) and four thicknesses (30, 50, 70, 100 mm). Participants rested randomly on one topper for 30 min. Measurements included pressure distribution, body temperature, EMG activity, and spinal alignment in lateral decubitus	The 30-mm topper reduced body pressure except at the biceps femoris region, improved lumbar alignment, and reduced body temperature. Harder toppers improved thoracic alignment. The authors suggested a 30-mm soft topper with a 100-mm hard topper limited to the thigh region
Normand et al. [64]	Quasi-experimental study	To analyze six experimental conditions using three support surfaces: no mattress, 8-cm foam, and 14-cm medium-density latex mattress	Level of evidence: III 10 asymptomatic participants. Tekscan pressure sensors recorded peak pressure over the chest, hip, and low back for 30 s in the supine position	Low-back support improved pressure distribution
Chen et al. [50]	Randomized crossover, single-blind controlled trial	To evaluate the effect of mattress stiffness on body-contact pressure and sleep comfort	Level of evidence: II 16 asymptomatic men aged 20–45 years. Supine and lateral positions were tested. Mattress types included plank springs, palm-fiber support with pillow top, a 3D foam-rubber/plant-fiber structure, and independent springs. Methods included ABW pressure measurement, ALICE polysomnography, and yes/no questionnaires	Sleep quality was optimized at an intermediate level of body-contact pressure
Denninger et al. [26]	Design process and simulation-validation study	To present body-mass distribution and compression forces exerted on the spine during lateral decubitus	Level of evidence: VI 3 participants (1 woman, 2 men). A custom mattress of extra-firm PU foam cubes with hollow ellipsoidal cavities was designed according to spinal curvature and body weight. Methods included camera acquisition, load cells, ANSYS finite-element modeling, and Optotrak optical measurements	The study proposed a model of human–mattress interaction. Customized cube cavities allowed load distribution with a mean deviation of 10%, and spinal alignment remained within 3% of shoulder-width distance
Van Deun et al. [58]	Repeated-measures, non-randomized controlled trial	To compare sleep quality with and without an active-control bedding system that independently adjusted firmness according to spinal alignment	Level of evidence: III 11 asymptomatic participants (5 women, 6 men). The Dynasleep mattress used sensors and adaptive spring pockets in inactive and active actuator modes. Measurements included body surface, sleep quality by polysomnography, spinal curvature, and subjective responses	Active-control mode improved sleep quality through adaptive support based on spinal needs, as shown by polysomnographic and questionnaire outcomes
Esquirol Caussa et al. [54]	Recommendation-model validation study	To analyze an automatic multimodal somatotype-determination model for selecting individualized mattress–pillow-topper combinations	Level of evidence: VI Pilot test: 6 participants. Second pilot test: 50 participants (28 women, 22 men; age 18–93 years; mean 34.43 years). Re-analysis: 117 participants (75 women, 42 men; age 4–93 years; mean 33.83 years). Supine position was assessed across 45 pillow–mattress combinations. Body dimensions and pressure distribution were measured using Kinect, tape, and pressure sensors	The model correlated with measured data: 85% for height and body circumference, 89.9% for weight, 80.4% for BMI, and 70% for morphotype categorization
Lee et al. [40]	Mixed factorial, non-randomized controlled trial	To evaluate sex-related differences in pressure distribution, perceived pain, and tolerance of the supine position	Level of evidence: III 10 healthy participants (5 women, 5 men), mean age 29.1 years. Participants lay supine on their own mattresses. Body–mattress pressure distribution and subjective feedback were collected using a pressure-measurement system, VAS, faces pain rating scale, and Iowa Pain Thermometer	The head region showed higher pressure and pain indices than the lower back region. Pain appeared after 10 min and progressively increased. The greatest sex-related differences were observed in back and pelvic-girdle pain
Lee et al. [51]	Repeated-measures, non-randomized controlled trial	To analyze differences in body-pressure distribution and pain when lying on the floor versus on a mattress	Level of evidence: III 10 healthy participants (5 women, 5 men), mean age 29.1 years. Participants lay in the supine position on the floor and on a mattress. Pressure distribution and subjective feedback were measured with a pressure-measurement system, VAS, faces pain rating scale, and Iowa Pain Thermometer	The head region showed high pressure. Pain was perceived throughout the body in both conditions, but leg pain was greater when lying on the floor and appeared after 1 min
Topiwala et al. [32]	Descriptive cross-sectional study	To examine the association between back pain, sleeping position, sleep quality, and sleep surface	Level of evidence: III 330 Indian college students aged 16–21 years completed a 3-month survey on height, weight, sex, sleep duration, sleeping posture, and physical activity	Back pain prevalence was higher among students sleeping on sofas (61.3%) or soft mattresses (64.7%) than among those using normal mattresses (57.6%). Sleeping approximately 8 h was associated with lower back-pain prevalence than shorter or longer sleep durations
Leilnahari et al. [52]	Design process, repeated-measures, non-randomized controlled trial	To present a customized mattress with zonal elasticity designed to maintain neutral spinal alignment and to compare spinal curvature on soft, firm, and customized mattresses	Level of evidence: III 25 male students in lateral position. Mattresses included soft polyurethane/memory foam, firm mattress, and custom-made mattress with zonal firmness based on a musculoskeletal model using PU and spiral pressure springs with different wire diameters. Spinal alignment was measured using cameras and BRG.LIFEMOD2007 modeling	Customized support provided the best vertebral alignment, followed by firm and then soft mattresses
López-Torres et al. [39]	Non-randomized controlled correlation study	To investigate differences in perceived comfort across mattress stiffness levels in young and older adults and to correlate subjective responses with objective measurements	Level of evidence: III 75 participants: 19 young adults and 56 older adults. Lateral, supine, and seated positions were tested on four mattresses with different firmness levels. Contact pressure and subjective perception were measured using a PLIANCE 19P pressure system and questionnaires	No relevant differences were found between young and older participants. Subjective feedback correlated with objective measurements. Comfort and ease of turning/getting up improved with increased firmness
Low et al. [41]	Randomized crossover, single-blind controlled trial	To analyze body-pressure distribution on different mattress materials in three sleeping positions	Level of evidence: II 20 young asymptomatic participants (10 women, 10 men). Lateral, supine, and prone positions were tested on a latex foam mattress and a high-density PU foam mattress. Pressure distribution was measured using TEKSCAN 5400N	The latex mattress reduced peak pressure and provided better pressure distribution than the PU mattress
Intolo et al. [22]	Randomized crossover trial	To compare an innovative latex mattress with a standard commercial latex mattress in terms of pain, muscle tension, sleep quality, satisfaction, and comfort	Level of evidence: III 120 volunteers without sleep disorders or spinal disease. All mattresses were medium firm. The commercial mattress had a flat surface; innovative mattresses had square-node designs with low, medium, or high node height for patients with back pain, healthy users, or older adults. Participants rested for 30 min in supine, prone, and side-lying positions. VAS and statistical analyses were used for comfort, pain, tension, and sleep quality	The innovative mattress reduced back pain (p = 0.028) and muscle tension and improved comfort and satisfaction (p < 0.05). The square-node design distributed body weight, reduced pressure peaks, and supported comfort across positions
Palmero et al. [42]	Recommendation-model validation study	To present a somatotype-determination model based on RGB-depth imaging and automatic reference-point extraction for mattress and pillow-topper recommendation	Level of evidence: III 200 participants (128 women, 72 men), age 4–93 years, mean 33.82 ± 23.02 years. Supine lying was assessed on an intermediate-density mattress. Measurements included body-surface contour and pressure distribution using Kinect and pressure sensors	The automated system showed higher accuracy than manual measurements for anthropometric categorization
Park et al. [55]	Design process, repeated-measures, non-randomized controlled trial	To introduce an adjustable bed that modifies the height of eight mattress zones and to compare perceived comfort on adjustable and flat-bed conditions	Level of evidence: III 64 healthy participants (35 women, 29 men), aged 25–50 years. Lateral, prone, and supine positions were tested. Contact pressure and subjective comfort in nine body regions were measured using a force-sensing resistor matrix and a 0–5 questionnaire	Height adjustments correlated with pressure distribution and subjective comfort. Participants preferred W-shaped adjustment in supine and lateral positions and U-shaped adjustment in prone/lateral-prone positions
Verhaert et al. [28]	Instrument design and validation study	To assess a method for estimating spinal curvature using a human model and mattress-indentation measurements	Level of evidence: III 65 participants (33 women, 32 men; mean age 27.3 ± 11.5 years) and a validation sample of 20 participants. Lateral, prone, and supine positions were tested on the Dynasleep mattress with active actuators or simulated sagging support. Body dimensions, body-surface contour, and spinal curvature were measured using calipers, tape, the IKELO optical system, and indentation sensors	Measured angular spinal deformation correlated well with estimated deformation (intraclass correlation 0.88–0.73)
Verhaert et al. [59]	Instrument design, validation, and recommendation-model study	To analyze spinal curvature using anthropometric models, bed-base properties, and mattress load-deflection characteristics and to identify optimal bed-base/mattress combinations	Level of evidence: III 18 participants (9 women, 9 men), mean age 28.5 ± 4.7 years. Nine combinations of bed base and mattress were tested in lateral position: homogeneous box spring, multi-zone slatted base, multi-zone mesh base; multi-zone pocket spring mattress, multi-zone latex mattress, and homogeneous PU foam mattress. Body surface and spinal curvature were measured with IKELO, zSnapper 3D scanning, and personalized modeling	The estimated and validated spinal curves showed approximately 85% correspondence
Verhaert et al. [60]	Mattress design process; randomized crossover, single-blind controlled trial	To evaluate an active-control mattress system capable of identifying body movement and sleep posture, estimating spinal shape, and adjusting regional firmness to promote neutral alignment	Level of evidence: III 18 participants (8 women, 10 men), mean age 31.3 ± 14.3 years; field study: 12 participants, mean age 38.7 ± 23.4 years. Sleep positions included supine, left/right lateral, prone, and intermediate positions. The Dynasleep mattress used built-in sensors and adaptive spring pockets. Spinal shape was estimated from indentation data and personalized human models	Sleep quality improved significantly when participants rested on the active-control mattress system
Rooh Ul Muazzam et al. [31]	Cross-sectional study	To examine the frequency of low back pain among young adults and its relationship with different mattress types	Level of evidence: III 366 Pakistani students (266 women, 100 men), mean age 22.06 ± 3.74 years. Inclusion required no spinal injury or deformity and sleeping on the same mattress for at least 3 months. Pain was measured using NPRS, and sleep-related information was collected using the MOS questionnaire	Low back pain was reported by 208 participants (56.4%); 105 of these had used the same mattress for more than 3 years. Sponge mattresses were frequently associated with symptoms, whereas spring mattresses appeared more favorable for pain reduction
Wu et al. [61]	Instrument design, repeated-measures study	To introduce a mattress-evaluation method based on back-surface contour, body-pressure distribution, and spinal shape in supine and standing positions	Level of evidence: III 17 healthy adults (4 women, 13 men), mean age 34.9 ± 9.7 years. Supine position was tested on a palm-fiber mattress and a bilayer latex/palm-fiber mattress. Measurements included back-surface contour, vertebral alignment/mattress indentation, and body-mattress contact pressure using 3D scanning, ANSYS finite-element modeling, and Tactilus pressure measurement	The bilayer latex/palm-fiber mattress better reproduced the standing back-surface contour in the supine position than the palm-fiber mattress alone
Yoshida et al. [56]	Correlation study: simulation versus subjective rating	To evaluate agreement between computer-simulation predictions and subjective comfort/preference ratings	Level of evidence: III 14 male students aged 21–24 years. A finite-element model was generated from three participants representing body-dimension coverage. Four pocket-coil mattresses were tested in supine position. Outcomes included internal stress, head and thorax displacement, and subjective ratings of sinking, firmness, and comfort	Subjective evaluations were consistent with simulation results, including von Mises stress in the cervical region and neck sinking
Zhong et al. [44]	Instrument design, validation, and mattress-design study	To predict spinal shape from mattress indentation and assess optimal zonal mattress stiffness	Level of evidence: III 9 women divided into three BMI groups. Fourteen mattress configurations were obtained by varying five regional stiffness zones using six spring-stiffness types. Mattresses included a PU foam surface layer and pocket-spring core. Vertebral curvature was measured using a custom indentation bar built into the mattress	Mean absolute and relative errors between estimated and experimental measurements were 3.4 mm (SD 2.7) and 9.27%. Lower-back and pelvic stiffening improved cervicothoracic, thoracolumbar, and lumbosacral alignment; shoulder alignment varied with hip and upper-back stiffness
Hong et al. [23]	Randomized crossover trial	To evaluate relationships among body-mattress interaction, stiffness, pressure distribution, intervertebral-disc stress, spinal curvature, and musculoskeletal health	Level of evidence: III 16 healthy students without spinal injury; mean age 29.4 years, mean height 170.3 ± 7.8 cm, mean weight 64.8 ± 13.6 kg, mean BMI 22.2 ± 3.4. Supine position was tested on three polymeric foam mattresses with different load deflection values: soft (20 lb), medium (42 lb), and hard (120 lb). Spinal curvature was measured using accelerometer-based curvature tape	Hard mattresses decreased lumbar lordosis and increased contact/peak pressure, potentially causing discomfort. Medium and hard mattresses supported the chest and pelvis better than the soft mattress, but excessive firmness may increase pressure-related symptoms
Bolton et al. [27]	Crossover trial	To investigate whether mattress zoning reduces spinal curvature and low back pain and improves sleep quality, and to assess associations with position, nocturnal movement, and skin temperature	Level of evidence: III 16 adults aged 18–70 years with chronic low back pain on at least 5 nights/week for at least 3 months. Exclusion criteria included pregnancy, inflammatory disease, fibromyalgia, cancer, scoliosis, spinal injury, use of multiple beds, and sleep/anti-inflammatory medication. Participants completed a 5-night protocol comparing their own bed and an experimental zoned mattress with interchangeable steel-spring blocks. NPRS ratings, sleep position, nocturnal movements, and room/skin temperature were recorded	The experimental mattress significantly reduced mean spinal curvature angle (p < 0.01) and improved comfort and pain after the fourth night. Participants slept approximately 7 h 54 min, mainly in lateral position, and moved approximately every 2.2 h
Yu et al. [11]	Repeated-measures experimental study	To evaluate the effect of an inflatable air mattress with variable rigidity on subjective and objective sleep quality	Level of evidence: III Healthy male participants. Different air-cell pressure conditions were tested. Subjective sleep quality and objective sleep-stage information were assessed using movement and brain-wave data	Variable rigidity adjusted to participant preference improved objective sleep-quality parameters, although subjective sleep ratings were not uniformly improved
Wei et al. [8]	Multi-night polysomnography study	To investigate the influence of an adjustable zoned air mattress on sleep parameters and possible sex-related differences	Level of evidence: III Healthy adult participants. A multi-night comparison was performed using polysomnography. Sleep parameters, EEG spectrum, heart rate, and mattress-related responses were assessed	The adjustable zoned air mattress influenced sleep parameters; female participants showed greater sensitivity to mattress-related changes
Price et al. [24]	Prospective A–B pilot study	To assess changes in sleep quality after use of an air-filled overlay mattress in patients with chronic pain	Level of evidence: III 19 patients with chronic low back pain and sleep disorders completed a 4-week study using a low-pressure inflatable overlay mattress. Subjective sleep comfort and pain were assessed	The air-filled overlay mattress produced statistically significant improvements in sleep quality and pain relief

ABW Actual Body Weight, ALICE Alice PDx Diagnostic System, BMI Body Mass Index, COBRA Danish low back pain questionnaire assessing pain and activities of daily living, EEG Electroencephalogram, EMG Electromyography, MOS Medical Outcomes Study, PU Polyurethane, NPRS Numeric Pain Rating Scale, REM Rapid Eye Movement, RGB Red, Green, Blue, RMS Root Mean Square, VAS Visual Analogue Scale

Eligibility assessment and methodological appraisal were performed by two independent reviewers. Disagreements were resolved by a third reviewer. Because of marked heterogeneity in study design, mattress materials, follow-up duration, populations, and outcome measures, a formal meta-analysis was not considered appropriate; therefore, results were synthesized qualitatively.

Results

A total of 47 studies met the inclusion criteria and were included in the review (Table 1). The analysis showed a consistent relationship between mattress characteristics, spinal alignment, perceived comfort, and low back pain. The evidence can be grouped into three domains: firmness level and mechanical support; pressure distribution and individualized or zonal systems; and mattress microclimate and technological innovation.

Across the 25-year period, the most consistent clinical finding was the superiority of medium-firm mattresses over very firm or very soft alternatives. Kovacs et al. [21], in a large randomized, double-blind, controlled multicenter trial, demonstrated that medium-firm mattresses were associated with greater improvement in chronic nonspecific low back pain and disability than firm mattresses. Similarly, Jacobson et al. [20] reported improved sleep quality and reduced back pain after the introduction of medium-firm bedding systems. These findings indicate that a mattress should not simply be hard to be clinically useful; rather, it should provide enough compliance to reduce pressure peaks while maintaining sufficient resistance to preserve spinal alignment.

Biomechanical and ergonomic studies provide a plausible explanation for this observation. Hard surfaces may increase focal pressure over the scapular, pelvic, and sacral regions, while excessively soft surfaces may permit excessive sinking of the pelvis or trunk. Medium-firm configurations appear to offer a better balance between pressure redistribution and preservation of lumbar lordosis or neutral lateral alignment [22, 23]. Several studies using body-pressure mapping, spinal-curvature assessment, electromyography, or finite-element approaches supported the importance of interface pressure and spinal geometry in determining comfort and musculoskeletal symptoms [15, 23, 25, 28, 44, 46, 56, 61].

Specialized overlay systems, air-filled mattresses, zonal support, and customized designs showed potential benefits in selected contexts. Air-adjustable or active-control systems were reported to reduce pain, improve sleep quality, or improve spinal alignment in some studies [18, 24, 58, 60]. Zonal systems, including mattresses with variable stiffness or different support areas for the shoulders, waist, and pelvis, may be clinically relevant because the human body does not load the mattress uniformly. Studies evaluating customized or adaptive support suggested that tailored systems may improve comfort and spinal alignment compared with uniform surfaces [26, 27, 52, 54, 59].

Material composition also influenced outcomes. Latex and selected foam mattresses were often associated with improved pressure distribution and perceived comfort, whereas spring systems provided robust mechanical support and may facilitate changes in position during sleep [30, 36, 41]. However, the available evidence does not support the superiority of one material in all patients. The clinical effect depends on the interaction between material density, surface design, body weight, sleeping posture, and user preference. Mattress material should therefore be interpreted together with firmness and support geometry rather than as an isolated determinant of efficacy.

The mattress microclimate emerged as a relevant but sometimes underappreciated factor. Studies on temperature regulation and thermal comfort indicated that skin temperature, heat capacity, and the ability of the bed system to facilitate thermoregulation may influence sleep depth and continuity [29, 37, 47, 63]. High-heat-capacity mattresses and temperature-control strategies have been associated with changes in slow-wave sleep and subjective sleep comfort, suggesting that restorative sleep depends not only on spinal support but also on thermal interaction between the body and the sleep surface.

Within the prespecified 1999–2023 search window, smart- and active-control mattress systems were identified as emerging technologies capable of adapting support to the user’s posture or pressure distribution [4, 11, 58, 60]. These systems may be relevant for patients whose pain varies according to position or whose anthropometric characteristics make standard mattress recommendations less reliable. However, large-scale clinical validation in patients with chronic low back pain remains limited. A 2025 review on home sleep-monitoring technologies was not included among the 47 studies because it fell outside the prespecified search window; it was retained only as contextual evidence on recent developments in smartphone applications, smartwatches, and smart mattresses [49].

Discussion

The findings of this 25-year systematic review converge on the principle that mattress selection is a clinically relevant determinant of sleep quality and spinal health. The traditional recommendation of an “orthopedic” or very firm mattress is not supported by the available evidence as a universal rule. Instead, literature most consistently supports medium-firm mattresses as the best general compromise between pressure relief and structural support [20, 21, 48]. This conclusion is important because many patients still interpret firmness as equivalent to therapeutic efficacy, whereas the data indicate that excessive firmness may be as problematic as excessive softness.

The primary mechanism through which mattress firmness influences low back pain is the maintenance of neutral spinal alignment during sleep. In the lateral decubitus position, the shoulders and pelvis must sink enough to avoid high-pressure peaks but not so much that the spine assumes a curved or rotated posture. In the supine position, the mattress must support the lumbar region while allowing distributed contact across the trunk and lower limbs. Studies using spinal-curvature measurements, pressure mapping, and computational models indicate that inappropriate support may increase intervertebral stress, muscle activity, and stiffness on awakening [23, 27, 52]. The clinical superiority of medium-firm mattresses observed in controlled trials is therefore biomechanically plausible.

A key point for clinical practice is that mattress efficacy is not independent of individual patient characteristics. Body mass index, body shape, sex-related pelvic and shoulder morphology, age, and preferred sleeping position modify the interaction between the body and the mattress. Individuals with higher body mass may experience excessive sinking even on surfaces perceived as medium firm, while underweight individuals may experience localized pressure pain on surfaces that are well tolerated by others. Female users may have different pressure-distribution requirements because of pelvic morphology, and older adults may need additional pressure relief because of skin fragility, joint stiffness, and reduced mobility. These factors explain why a single mattress type cannot be prescribed with absolute certainty for all patients.

The evidence on zonal, air-adjustable, and customized mattresses is promising because these systems attempt to adapt support to regional anatomical needs. Mattresses with differentiated support zones may better accommodate the shoulders, waist, and pelvis, while active-control systems may adapt to postural changes during the night. Such technologies are conceptually attractive for patients with recurrent low back pain, restless sleep, or marked anthropometric asymmetry. Nevertheless, the clinical evidence remains preliminary, and many studies have small samples, short follow-up, or healthy volunteer populations. Future trials should include symptomatic patients, standardized pain and disability outcomes, longer follow-up, and clear reporting of mattress properties.

Thermal microclimate should also be integrated into mattress recommendations. A sleep surface that provides appropriate spinal support but retains excessive heat may still impair sleep quality. Conversely, materials that facilitate heat exchange and maintain a comfortable skin temperature may improve sleep continuity and slow-wave sleep [29, 37, 47]. This aspect is especially relevant in older adults, individuals with insomnia symptoms, and patients whose pain perception is amplified by fragmented sleep. The best mattress should therefore be considered as a biomechanical and thermoregulatory system, not merely as a support surface.

This review has several limitations. First, studies differed substantially in design, sample size, mattress classification, intervention duration, and outcome measures. Second, many studies used subjective comfort scores, which are clinically meaningful but difficult to compare across protocols. Third, mattress firmness was often described using commercial or qualitative terminology rather than standardized mechanical measurements. Fourth, several studies evaluated healthy participants, limiting direct generalizability to patients with chronic low back pain. Fifth, blinding is difficult in mattress trials because participants may perceive surface differences, which may influence subjective outcomes. Finally, the review was not prospectively registered, and this should be considered when interpreting the findings.

Despite these limitations, the evidence is sufficiently consistent to support practical recommendations. For adults with nonspecific low back pain or morning stiffness, a medium-firm mattress should be considered the most reasonable initial choice. The mattress should maintain neutral spinal alignment, distribute pressure over the shoulders and pelvis, facilitate position changes, and provide a comfortable thermal microclimate. In patients with atypical body habitus, persistent symptoms, older age, or specific pressure intolerance, zoned or adjustable systems may be considered, although the recommendation should remain individualized and cautious. Smart mattresses and artificial-intelligence-driven surfaces represent a promising field but should not yet replace clinically validated principles of support, pressure redistribution, and patient-specific assessment.

Future research should prioritize high-quality randomized controlled trials with standardized definitions of mattress firmness, objective measurements of body pressure and spinal alignment, validated pain and disability scales, and longer follow-up periods. Studies should stratify outcomes by body mass index, age, sex, sleep posture, baseline pain severity, and mattress age. Such data would allow clinicians to move from general advice toward more precise, individualized mattress prescriptions.

Conclusions

Over the past 25 years, available literature has progressively shifted away from the simplistic concept that a very hard “orthopedic” mattress is the best option for spinal health. The most consistent evidence supports medium-firm mattresses as a practical and clinically reasonable choice for improving sleep quality and reducing low back pain or morning stiffness. Their benefit appears to depend on the ability to preserve neutral spinal alignment while distributing pressure evenly across the shoulders, trunk, pelvis, and lower limbs.

The effectiveness of medium-firm support can be enhanced by appropriate surface design, topper characteristics, zonal support, and attention to thermal comfort. Because body habitus, age, sex-related anthropometry, sleep posture, and personal comfort all influence the body–mattress interaction, mattress recommendations should be individualized whenever possible. Technological advances, including adjustable and smart mattresses, may further improve personalized sleep support, but stronger clinical validation is required before they can be considered standard care. In current practice, clinicians should recommend mattresses that combine medium-firm support, pressure redistribution, neutral spinal alignment, and a comfortable microclimate, while avoiding unsupported claims based solely on the term “orthopedic.”

Abbreviations

ADL

Activities of daily living

BMI

Body mass index

CASP

Critical Appraisal Skills Programme

CBT

Core body temperature

EMG

Electromyography

HHCM

High-heat-capacity mattress

LBP

Low back pain

LEGEND

Literature evaluation and grading of evidence

NPRS

Numeric pain rating scale

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PU

Polyurethane

REM

Rapid eye movement

RMS

Root mean square

SWS

Slow-wave sleep

VAS

Visual analog scale

Authors’ contributions

C.G., P.F., P.S., A.A., F.A., C.C., P.L., and D.C., made substantial contributions to design, analysis, and interpretation of data and contributed to the writing and revision of the manuscript. D.G, C.E., and P.I., collaborated with the writing and revision of the manuscript. All authors read and approved the final manuscript.

Funding

This research was funded by the European Union under the Next Generation EU iniziative - NRRP M6C2, Investment 2.1: Enhancement and Strengthening of Biomedical Research in the NHS (project no. PNRR-MAD-2022-12376692, CUP H83C22000940001).

Availability of data and materials

Studies carried out between 1999 and 2023 have been included in the current review. The mattresses examined in the selected studies were assessed in terms of firmness, composition, and technological innovation, with a particular focus on their impact on spinal alignment and sleep quality. All authors read and approved the final manuscript.

Declarations

Ethics approval and consent to participate

No formal ethical approval was required for this study. All data analyzed were extracted exclusively from previously published research. No individual patient information or unpublished material was accessed or used.

Consent for publication

All authors consent to the publication of this manuscript.

Authors’ information

Not applicable.

Competing interests

The authors declare no competing interests.