Insecure attachment style predicts low bone mineral density in postmenopausal women. A pilot study
Lo stile di attaccamento insicuro è un fattore di rischio di ridotta densità
minerale ossea in donne in menopausa. Uno studio pilota

3, MICHELE RIBOLSI1, Adalgisa pietropolli3, carlo ticconi3,

1Chair of Psychiatry, Department of Systems Medicine, University of Rome “Tor Vergata”, Rome, Italy
2Psychiatric Clinic, Fondazione Policlinico “Tor Vergata”, Rome, Italy
3Section of Ginecology and Obstetrics, Department of Biomedicine and Prevention, University of Rome “Tor Vergata”, Rome, Italy.
4Chair of Orthopaedics and Traumatology, Department of Surgery, University of Rome “Tor Vergata”, Rome, Italy
5Orthopaedic Clinic, Fondazione Policlinico “Tor Vergata”, Rome, Italy

SUMMARY. Introduction. Major depressive disorder (MDD) and osteoporosis are two common disorders with high morbidity and mortality rates. Conflicting data have found associations between MDD and low bone mineral density (BMD) or osteoporosis, although causative factors are still unclear. A pilot study was designed with the aim to assess the relationship between MDD and BMD in postmenopausal women with MDD compared to healthy volunteers. We hypothesized that attachment style (AS) mediated this relationship. Methods. The sample was made of 101 postmenopausal women, 49 with MDD and 52 age-matched healthy volunteers. Structured clinical interview and Beck Depression Inventory (BDI) were performed to assesse MDD. AS was evaluated using the Relationship Questionnaire (RQ). BMD was measured by dual energy X-ray absorptiometry. Results. The univariate analysis showed that women with MDD had lower BMD values as compared to healthy volunteers. In the regression models MDD diagnosis and BDI score were not significant predictors of low BMD. The “preoccupied” pattern of insecure AS was a significant, independent predictor of decreased BMD in all skeletal sites: lumbar spine (p=0.008), femoral neck (p=0.011), total hip (p=0.002). Conclusion. This is the first study exploring the relationship between AS, MDD and BMD. Our results support the link between MDD and low BMD. We found that insecure AS was a risk factor for decreased BMD, regardless of depression. Insecure AS may play a role in the relationship between MDD and BMD or may constitute a risk factor itself. Therapeutic interventions focused on AS could improve psychiatric disorders and physical diseases related to low BMD.

KEY WORDS: attachment style, bone mineral density, major depressive disorder, osteoporosis, postmenopausal women.

RIASSUNTO. Introduzione. La depressione maggiore (MD) e l’osteoporosi sono malattie ad alta prevalenza nel genere femminile, associate a morbosità e mortalità. Sebbene alcuni studi abbiano dimostrato un’associazione tra MD, ridotta densità minerale ossea (BMD) e osteoporosi, non sono stati chiariti i meccanismi causali. Lo stile di attaccamento insicuro è stato messo in relazione con la patogenesi e il decorso di malattie croniche come la MD e le malattie cardiovascolari. Obiettivo di questo studio pilota è esplorare la relazione tra MD e BMD. Si ipotizza che lo stile di attaccamento possa agire da mediatore. Metodi. Il campione è formato da 101 donne in menopausa, 49 con MD e 52 controlli sani. La diagnosi di MD è stata formulata con l’intervista clinica e la Beck Depression Inventory. Lo stile di attaccamento è stato esplorato usando il Relationship Questionnaire, la BMD con la Mineralometria Ossea Computerizzata con tecnica DXA (Dual energy X-ray Absorptiometry). Risultati. L’analisi univariata ha mostrato che le donne con MD avevano valori di BMD inferiori rispetto ai controlli sani. Nelle analisi di regressione multipla la MD non è emersa come predittore significativo di ridotta BMD. Lo stile di attaccamento insicuro “preoccupato” è risultato un predittore significativo di ridotta BMD in tutti i siti scheletrici misurati con la DXA: colonna vertebrale lombare (p=0,008) e segmenti femorali: “femoral neck” (p=0,011), “total hip” (p=0,002). Conclusioni. Questo è il primo studio che esplora il possibile ruolo di MD e stile di attaccamento sulla BMD. Lo stile di attaccamento è risultato un predittore di ridotta BMD, indipendentemente dalla MD. L’attaccamento insicuro potrebbe avere un ruolo nella patogenesi dell’osteoporosi anche indipendente dalla MD. Se questi risultati saranno confermati, gli interventi terapeutici focalizzati sullo stile di attaccamento potrebbero contribuire al miglioramento della comorbilità psichiatrica e medica legata all’osteoporosi.

PAROLE CHIAVE: stile di attaccamento, depressione, menopausa, donne, osteoporosi, densità minerale ossea.

Major depressive disorder (MDD) and osteoporosis are two common disorders with high morbidity and mortality rates1. According to the World Health Organization (WHO), it has been estimated that by 2020, MDD will be one of the leading causes of years lived with a disability2. It is worth noting that this development is not only attributed to depression itself, approximately half of the patients with MDD have at least one co-morbid either psychiatric or medical illness3. Major depression has been associated with accelerated bone loss leading to the development of low bone mineral density (BMD) and osteoporosis4-6. Even though, some data investigating the effects of depression on BMD have yielded divergent results, accounting for the exclusion of MDD from recognized cause of osteoporosis7. Given that 20%-30% of postmenopausal women with osteoporosis have a secondary cause, diagnosing secondary osteoporosis is mandatory to provide appropriate interventions8. Two main factors might explain the failure in demonstrating a putative link between MDD and low BMD. The first of these factors is methodological. As clearly suggested by a recent meta-analysis, a significant association between MDD and lower BMD was found among women diagnosed by a psychiatrist, compared to self-reported screening questionnaires9. The second factor is about the etiology of the association between MDD and BMD. In subjects with MDD, the role of various causative factors of low BMD has been discussed. These include the hypothalamic-pituitary-adrenal (HPA) axis hyperactivity10, imbalance of inflammatory cytokines11, impairment in gonadal hormones levels, life-style12, and antidepressant intake13. In this study we aimed at investigating the possible contribute of further psychiatric variables in determining the interplay within MDD and decreased BMD. In particular, we focused on adult attachment style (AS). Adult attachment style is conceived as an attribute of the personality arisen from early life experiences with caregivers and representing the style of relating in adult close relationships. According to AS theory14, people differ in the degree to which they believe close others will be supportive and available in times of need. Therefore, through previous relationships, people develop different “internal working models” of their close relationships that include sets of expectations, beliefs, and desires about whether the other will be responsive and whether the self is worthy of love15. These distinct working models lead by example in adult relationships and help individuals manage daily life stressors16. A growing literature made out the association between AS and MDD. A secure AS is needed for healthy development17. Insecure AS is an established factor that underlies the development and the course of MDD18. Furthermore, AS has been linked to the development of different diseases and non-communicable diseases, such as cardiovascular diseases, physical pain, and gastroduodenal ulcers19. AS may influence the course of chronic diseases by means of biological pathways, attitudes to support-seeking, and other illness-related behaviors20,21. To the best of our knowledge, there are no published studies investigating a potential link between AS and BMD. The present pilot study aimed at confirming whether MDD is a risk factor for low BMD among postmenopausal women, and exploring the mediating role of insecure AS.
From September 2012 to September 2013, postmenopausal women with a MDD diagnosis from the outpatient program at “Tor Vergata” University Psychiatry Clinic were considered for the study. Inclusion criteria were a current depressive episode and being less than 70 years of age. Exclusion criteria were a history of eating disorders and, in the six months prior to the study, a history of alcohol and/or substance abuse. Postmenopausal HV of comparable ages and socio-economic status without a personal and familial history of psychiatric disorders were recruited as controls. To counter a potential confounding effect of bone turnover, three additional exclusion criteria in both groups were introduced: a history of rheumatoid arthritis, vitamin D deficiency 22, and chronic use of corticosteroids. Of the 101 women enrolled in the study, the 49 affected by a current episode of MDD comprised the MDD group and the 52 healthy postmenopausal women comprised the HV group. All women with MDD were medicated with SSRIs.
All study procedures and protocols were approved by the University Intramural Ethical Committee. Participants were provided written and oral information about the study prior to enrollment and were informed that they could withdraw from the study at any point, without detriment.
All participants were administered an accurate, semi-structured clinical interview in order to identify predictors of low BMD, including current smoking habit, diagnosed osteoporosis, type 1 diabetes, hyperthyroidism, and antidepressant drug therapy in the six months prior to the study.
Experienced clinical psychiatrist administered the Mini-Plus International Neuropsychiatric Interview (MINI-Plus) to verify the diagnosis of MDD or the absence of psychiatric disorders23.
The Beck Depression Inventory (BDI) was administered to measure the severity of current depressive symptoms. BDI is a 21 item self-report instrument using a four-point scale ranging from 0 (symptom not present) to 3 (symptom very intense). Women were asked to place a mark next to the statement best describing how they felt over the week prior to the study for each of the 21 items24.
Adult AS was assessed using the Relationship Questionnaire (RQ). Participants were instructed to interpret the questionnaire in reference to all their close relationships with peers (whether romantic or not). The RQ is a single-item measure comprising four short paragraphs, each describing a prototypical attachment pattern as it applies in close adult peer relationships. For each of the four descriptions, the respondents indicate how well it describes or relates to themselves on a seven-point rating scale. RQ provides a four-category model of AS based on the four combinations obtained by dichotomizing the subject’s mental representations of the self (self “internal working model” on one axis) and the subject’s image of the other (other “internal working model” on the orthogonal axis) into “positive” and “negative,” based on their interpersonal relationships. This yields four attachment patterns: secure (positive self, positive other), preoccupied (negative self, positive other), fearful (negative self, negative other), and dismissing-avoidant (positive self, negative other) 25-27.
BMD was measured by Dual energy X-ray absorptiometry (DXA) at three skeletal sites, lumbar spine (L1-L4) and the proximal femur (femoral neck, total hip). According to WHO, DXA is the gold standard method of measuring BMD and has widespread use because of its reliability, non-invasiveness, minimal radiation exposure, and short-term procedure preparation requirements. BMD determination is the main tool for assessing osteoporosis.
Statistical analysis
Since the DXA indices showed a non-normal distribution, we logarithmically transformed these variables to obtain a better approximation to a Gaussian curve, achieving appropriate equivalence to a normal distribution (Kolmogorov-Smirnov test, p>0.2). We conducted t-tests and chi-square tests to compare differences between HV and MDD groups. Pearson’s product-moment correlation analysis was performed for DXA indices and age, years since menopause, RQ scales, and total BDI score for the whole sample, HV-, and MDD group. Finally, multiple forward stepwise regression analysis was used to identify significant independent predictors of DXA indices by selecting independent variables among those that were found to be significant in the univariate and correlation analyses. Statistical significance was set at p<0.05.
Descriptive and univariate analysis of socio-demographic and clinical characteristics are reported in Table 1. Compared to the HV group, the group with MDD had a higher total BDI score and lower BMD values.

Correlation analysis
In the whole sample, significant negative correlations were found between “fearful” RQ scale and BMD values of the three skeletal sites: lumbar spine (p=0.008), femoral neck (p=0.019), and total hip (p=0.017). We also found significant negative correlations between “preoccupied” RQ scale and BMD values of two of the three skeletal sites: femoral neck (p=0.013), and total hip (p=0.024) among all participants. The correlation analysis yielded different results when MDD and HV groups were analyzed separately. For the MDD group, a significant negative correlation was found between “fearful” RQ subscale and femoral neck (p=0.008) and total hip BMD values (p=0.031) and between the “dismissing” RQ subscale and femoral neck BMD values (p=0.023). No significant correlations were found for HV group. Furthermore, the relationships between BMD values and MDD onset, as well as between duration and episode number were not significant (Table 2).

Regression analysis
We developed three regression models, one for each region of BMD measurement (dependent variable). Years since menopause, total BDI score, “preoccupied,” “fearful,” and “dismissing” RQ scales, and the grouping factor “HV vs. MDD” were entered into the regression models as independent variables. Results of the regression analyses are reported in Table 3. The regression models were able to significantly predict BMD values for spine (L1–L4), femoral neck, and total hip, explaining 6% to 18% of the variance. The “preoccupied” RQ scale was an independent significant predictor in the three regression models. Years since menopause emerged as significant predictor in the femoral neck and total hip regression models. Other variables did not significantly predict BMD values in the three regression models.

Our results showed that the group of women with MDD had lower BMD values compared to the HV. No differences were found between the two groups for other established risk factors of low BMD, such as family history of osteoporosis, thyroid diseases, hypertension, diabetes, and smoking. Afterwards, the results of the present study support the hypothesis that MDD represents an important though often disregarded risk factor for osteoporosis.
In this study, we adopted a rigorous recruitment procedure. Diagnosis of MDD included but was not restricted to self-report questionnaire28. For instance, we enrolled a sample of women whose clinical diagnosis was first developed through a self-report scale (the BDI) and then confirmed by a semi-structured interview (the MINI-Plus) conducted by the same psychiatrist. Moreover, we employed a sample of postmenopausal women, HV and with MDD, with the intent to minimize possible confounding factors such as hormonal, nutritional, and physical ones, which might be different before and after menopause29.
The correlation analysis demonstrated that women with insecure AS, with preoccupied and fearful patterns, showed decreased BMD at the three skeletal sites (lumbar spine, femoral neck, and total hip). However, when analyzing the MDD group and the HV group separately, we found a significant correlation between insecure AS and BMD in the MDD group but not in the HV group. This result supported our hypothesis, that insecure AS influenced BMD only in women with MDD.
Finally, we wanted to determine the most suitable predictors of decreased BMD in the whole sample of the study. The regression models for the sample included the following variables as possible risk factors for decreased BMD: years since menopause, the four AS subscales (secure, preoccupied, fearful, and dismissing), current depression severity (measured by BDI total score), and the “HV vs. MDD” grouping variable. Surprisingly, both MDD diagnosis and depression severity, were not significant predictors of low BMD. We found instead that age and insecure AS were significant and independent predictors of low BMD. Therefore, insecure AS but not depression was found to be a risk factor for low BMD. This result pointed out that neither depressive symptoms nor MDD influenced BMD values. Older postmenopausal women with insecure AS were more likely to show a decreased BMD at all skeletal sites.
At this stage, we can speculate about pathogenic pathways of low BMD in women with insecure AS. For instance, AS system is a psychological framework contributing to the stability and the flexibility of individual’s participation in affective relationships30. In this perspective, due to the fact that AS is pervasively involved in all social situations, insecure AS may specifically, and perhaps more than secure AS31, contribute to a wider range of pathogenic processes involved in reduced BMD. Individuals with insecure AS are deeply affected by their significant relationships, which evoke high levels of anxiety and fears of being abandoned32. Accordingly, it has been suggested that the way in which partners perceive and provide feeling to each other influences their cortisol responses33. A recent study supported the hypothesis that these links are different in women compared to men34. Data had consistently shown an increase in cortisol levels among individuals with insecure AS35. Therefore, those with insecure AS easily experienced social interactions as stressful, resulting in the activation of a chronic hyper-surveillance mechanism to deal with the intimidating environment, as perceived by them36-38. Through epigenetic pathways, this process could potentially modulate HPA axis activity, a principal stress-response system in humans, leading to the sustained release of cortisol39. Glucocorticoid is a well-documented cause of secondary osteoporosis, both in animal and in human models40. In individuals with insecure AS, HPA axis hyperactivity and imbalanced cortisol levels may cause changes in bone cell activity that affect expected bone turnover through the suppression of bone formation and the increase of bone resorption41. This cascade of events may result in decreased BMD. Further studies, including blood tests or saliva tests for the measure of cortisol levels and the possible HPA axis imbalance, are needed in order to understand the role of insecure AS on BMD, and to examine other factors influencing bone metabolism.
We acknowledge a main limitation of this study. Previous researches reported the association between antidepressant therapy and low BMD, independently from the effect of depression42,43. Antidepressant therapy could have been a confounding variable that we did not account, influencing the relation between MDD and BMD.
Our results suggest that insecure AS is a risk factor for lower BMD, regardless of depression. Attachment style, a psychological variable involved in all human relationships, may play a significant role in the complex interplay between depression and osteoporosis or may constitute a risk factor itself. Confirming or elucidating the influence of insecure AS could help us to comprehend the relationship between depression, AS, and BMD or osteoporosis. As osteoporosis and MDD are widespread disabling disorders, therapeutic interventions targeting insecure AS may further reduce the burden of these two diseases 44.
If the relationship between insecure AS and low BMD will be confirmed, we might expect that these therapeutic programs could be effective not only in improving psychiatric disorders but also physical diseases related to reduced BMD45.

Conflict of interest: the authors declare that have no conflict of interest.
The authors declare that the current study was conducted in conformity with the Helsinki Declaration concerning human rights and informed consent was obtained from each participant.
 1. Sattui SE, Saag KG. Fracture mortality: associations with epidemiology and osteoporosis treatment. Nat Rev Endocrinol 2014; 10: 592-602.
 2. Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990-2020: Global burden of disease study. Lancet 1997; 349: 1498-504.
 3. Bhattacharya R, Shen C, Sambamoorthi U. Excess risk of chronic physical conditions associated with depression and anxiety. BMC Psychiatry 2014; 14: 10.
 4. Calarge CA, Butcher BD, Burns TL, Coryell WH, Schlechte JA, Zemel BS. Major depressive disorder and bone mass in adolescents and young adults. J Bone Miner Res 2014; 29: 2230-7.
 5. Cizza G. Major depressive disorder is a risk factor for low bone mass, central obesity, and other medical conditions. Dialogues Clin Neurosci. 2011; 13: 73-87.
 6. Zong Y, Tang Y, Xue Y, et al. Depression is associated with increased incidence of osteoporotic thoracolumbar fracture in postmenopausal women: a prospective study. Eur Spine J 2015 May 23. [Epub ahead of print].
 7. Yazici AE, Bagis S, Tot S, Sahin G, Yazici K, Erdogan C. Bone mineral density in premenopausal women with major depression. Joint Bone Spine 2005; 72: 540-3.
 8. World Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Report of a WHO Study Group. World Health Organ Tech Rep Ser 1994; 843: 1-129.
 9. Cizza G, Primma S, Coyle M, Gourgiotis L, Csako G. Depression and osteoporosis: a research synthesis with meta-analysis. Horm Metab Res 2010; 42: 467-82.
10. Yirmiya R, Goshen I, Bajayo A, et al. Depression induces bone loss through stimulation of the sympathetic nervous system. Proc Natl Acad Sci U S A 2006; 103: 16876-81.
11. Alesci S, Martinez PE, Kelkar S, et al. Major depression is associated with significant diurnal elevations in plasma interleukin-6 levels, a shift of its circadian rhythm, and loss of physiological complexity in its secretion: clinical implications. J Clin Endocrinol Metab 2005; 90: 2522-30.
12. Cizza G, Ravn P, Chrousos GP, Gold PW. Depression: a major, unrecognized risk factor for osteoporosis? Trends Endocrinol Metab 2001; 12: 198-203.
13. Chau K, Atkinson SA, Taylor VH. Are selective serotonin reuptake inhibitors a secondary cause of low bone density? J Osteoporos 2012; 2012: 323061. doi: 10.1155/2012/323061.
14. Bowlby J. Attachment vol. 2: Separation and loss. New York, NY: Basic Books, 1973.
15. Puig J, Englund MM, Simpson JA, Collins WA. Predicting adult physical illness from infant attachment: a prospective longitudinal study. Health Psychol 2013; 32: 409-17.
16. Rees CA. Thinking about children’s attachments. Arch Dis Child 2005; 90: 1058-65.
17. Abdul Kadir NB, Bifulco A. Insecure attachment style as a vulnerability factor for depression: recent findings in a community-based study of Malay single and married mothers. Psychiatry Res 2013; 210: 919-24.
18. Fowler JC, Allen JG, Oldham JM, Frueh BC. Exposure to interpersonal trauma, attachment insecurity, and depression severity. J Affect Disord 2013; 149: 313-8.
19. McWilliams LA, Bailey SJ. Associations between adult attachment ratings and health conditions: evidence from the National Comorbidity Survey Replication. Health Psychol 2010; 29: 446-53.
20. Ciechanowski P, Sullivan M, Jensen M, Romano J, Summers H. The relationship of attachment style to depression, catastrophizing and health care utilization in patients with chronic pain. Pain 2003; 104: 627-37.
21. Ponizovsky AM, Drannikov A. Contribution of attachment insecurity to health-related quality of life in depressed patients. World J Psychiatry 2013; 3: 41-9.
22. Frighi V, Morovat A, Stephenson MT, White SJ, Hammond CV, Goodwin GM. Vitamin D deficiency in patients with intellectual disabilities: prevalence, risk factors and management strategies. Br J Psychiatry 2014; 205: 458-64.
23. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998; 59 Suppl 20: 22-33; quiz 34-57.
24. Beck A, Ward C, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961; 4: 561-71.
25. Carli L. Stili di attaccamento tra giovani adulti: analisi di un modello a quattro categorie. In: Attaccamento e rapporto di coppia. Milano: Raffaello Cortina Editore, 1995.
26. Bartholomew K, Horowitz LM. Attachment styles among young adults: a test of a four-category model. J Pers Soc Psychol 1991; 61: 226-44.
27. Scharfe E, Bartholomew K. Reliability and stability of adult attachment patterns. Pers Relationships 1994; 1: 23-43.
28. Siracusano A, Ribolsi M, Niolu C. [The 5th revised edition of the DSM: a revolution half-accomplished]. Recenti Prog Med 2014; 105: 141-3.
29. Yirmiya R, Bab I. Major depression is a risk factor for low bone mineral density: a meta-analysis. Biol Psychiatry 2009; 66: 423-32.
30. Niolu C, Siracusano A. Psychological issues in improving adherence and alliance. In: Sacchetti E, Vita A, Siracusano A, Fleischhacker W (eds). Adherence to antipsychotics in Schizophrenia. Milano: Springer, 2014.
31. Simpson JA, Rholes SW. Adult attachment orientations, stress, and romantic relationships. advances in experimental social psychology 2012; 45: 279-328.
32. Pietromonaco PR, DeBuse CJ, Powers SI. Does attachment get under the skin? Adult romantic attachment and cortisol responses to stress. Curr Dir Psychol Sci 2013; 22: 63-8.
33. Gunlicks-Stoessel ML, Powers SI. Romantic partners’ coping strategies and patterns of cortisol reactivity and recovery in response to relationship conflict. J Soc Clin Psychol 2009; 28: 630-49.
34. Laurent H, Laurent S, Hertz R, Egan-Wright D, Granger DA. Sex-specific effects of mindfulness on romantic partners’ cortisol responses to conflict and relations with psychological adjustment. Psychoneuroendocrinology 2013; 38: 2905-13.
35. Roque L, Veríssimo M, Oliveira TF, Oliveira RF. Attachment security and HPA axis reactivity to positive and challenging emotional situations in child-mother dyads in naturalistic settings. Dev Psychobiol 2012; 54: 401-11.
36. Jaremka LM, Glaser R, Loving TJ, Malarkey WB, Stowell JR, Kiecolt-Glaser JK. Attachment anxiety is linked to alterations in cortisol production and cellular immunity. Psychol Sci 2013; 24: 272-9.
37. Heim C. The dexamethasone/cotricotropin-relasing Factor in men with major depression: role of childhood trauma. Biol Psychiatry 2008; 63: 398-405.
38. Quirin M, Pruessner JC, Kuhl J. HPA system regulation and adult attachment anxiety: individual differences in reactive and awakening cortisol. Psychoneuroendocrinology 2008; 33: 581-90.
39. Pierrehumbert B, Torrisi R, Ansermet F, Borghini A, Halfon O. Adult attachment representations predict cortisol and oxytocin responses to stress. Attach Hum Dev 2012; 14: 453-76.
40. Bouvard E, Legrand E, Audran M, Chappard D. Glucocorticoid-induced osteoporosis: a review. Clinic Rev Bone Miner Metab 2010; 8: 15-26.
41. Rizzoli R, Cooper C, Reginster JY, et al. Antidepressant medications and osteoporosis. Bone 2012; 51: 606-13.
42. Fernandes BS, Hodge JM, Pasco JA, Berk M, Williams LJ. Effects of depression and serotonergic antidepressants on bone: mechanisms and implications for the treatment of depression. Drugs Aging 2016; 33: 21-5.
43. Wang Q, Chen D, Nicholson P, et al. The associations of serum serotonin with bone traits are age- and gender-specific. PLoS One 2014; 9: e109028.
44. Bruce J, Gunnar MR, Pears KC, Fisher PA. Early adverse care, stress neurobiology, and prevention science: lessons learned. Prev Sci 2013; 14: 247-56.
45. Krahé C, Paloyelis Y, Condon H, Jenkinson PM, Williams SC, Fotopoulou A. Attachment style moderates partner presence effects on pain: a laser-evoked potentials study. Soc Cogn Affect Neurosci 2015; 10: 1030-7.