Depressive disorders are a major source of non-fatal disease burden in developed countries, as well as being a key determinant of health-related disability in the developing world [1,2]. To achieve meaningful reductions in depression-related health burden a range of preventative and treatment strategies are urgently required [3,4]. The development of such initiatives, however, usually depends on broad recognition that depression is not only common and disabling but that it also responds to evidence-based treatments.

The concept of 'mental health literacy' has emerged to describe "knowledge and beliefs about mental disorders which aid their recognition, management or prevention" [5]. To date, assessment has rested largely on typical case-based vignettes of persons with schizophrenia or depression. This approach requires the respondent to identify the person as a 'case' of mental disorder and then disclose their knowledge or attitudes about available health services or treatments. This has proved to be a useful way of tracking community attitudes to a range of mental health problems and their treatments, particularly in English-speaking countries [6].

We propose an alternative approach which specifically describes depression as a health condition, and then seeks to determine understanding of its characteristics and impacts relative to other medical and psychological conditions. Through our social marketing strategies for beyondblue: the national depression initiative in Australia, we focused more on these general health and depression-specific literacy issues than on an individual's capacity to recognise a typical 'case' of mental disorder. Our earlier work indicated that while the general public was aware of the potential impacts of depression, common mental disorders generally were not seen to be as important as other major physical disorders such as cancer and heart disease [7]. Further, a range of other factors such as personal or family experiences of depression, previous experiences of seeking care or encountering other stigmatising attitudes towards persons with depression appeared to be impacting on willingness to seek mental health care.

As part of an international movement to promote greater recognition of the social and economic costs of depression [3,4], we were keen to develop this alternative approach to rating depression literacy. A priority was that the method should be able to be utilised easily in countries where English is not the first language or where there was little priority given to the provision of relevant health services. Consequently, we developed a modular depression literacy self-report instrument for use primarily in countries that are participating in the 'Reducing the Social and Economic Burdens of Depression' (SEBoD) movement in Asia.

As the SEBoD movement aims to improve depression literacy within general health workforces, we were interested to determine whether the instrument was sensitive to differences in mental health training. As many countries are not culturally homogeneous, we also sought to determine whether significant ethnic differences had impacts on depression literacy among persons residing in the same country. Therefore, in this Australian study we endeavoured to examine the utility of our instrument when it was completed by university students from different ethnic and health training backgrounds.

The study was a cross-sectional survey of four groups of Australian university students. There were two groups of general students (ethnic Chinese and non-ethnically selected students) and two groups of medical students (second and fourth year students of The University of Sydney Graduate Medical Program).

The English version of the International Depression Literacy Survey (IDLS) was administered to four student groups: medical students from The University of Sydney in second year (n = 103) and fourth year (n = 82); tertiary students from The ANU (n = 38); and Chinese students from The University of Sydney, The University of Technology (Sydney) and The University of New South Wales (n = 184). In total the survey was completed by N = 407 students. Eighty-two percent of the sample completed the entire survey (n = 333/407), 33 terminated after section B (i.e. only completed the first two sections), 17 after section C, 12 after section D, six after section E and six after section F.

The IDLS has been developed specifically to track understanding of depression as a general health condition. For the IDLS to be useful within ethnically diverse communities or groups of health professionals over time, it should first be sensitive to likely baseline differences between relevant groups residing in the same country. While there is little published research regarding the specifics of cross-cultural differences regarding attitudes and beliefs around depression, studies published to date suggest attitudes to mental illness in general differ between Asian and western cultures [18,19]. The study reported here demonstrates that the IDLS does detect clear differences between medical students in second and fourth year courses, and between non-medical students from ethnic Chinese backgrounds and other undergraduates residing in Australia. These differences are obvious across all the key areas (depression within the context of major general health problems and major mental health problems, common psychological symptoms of depression, attitudes to the use of evidence-based treatments, patterns of health care utilisation and expectations of discrimination).

The basic utility of the instrument for potential use in wider epidemiological, cross-cultural, longitudinal or interventional studies is given preliminary support by the current results. However, it must be noted that convenience samples were utilised in this study, thus limiting the conclusions that can be drawn. The effect of health training on responses would ideally be confirmed by a longitudinal study, sampling the same individuals prior to and following mental health training. Another limitation was the age difference within each of the comparison student groups. While there did not appear to be a systematic effect with increasing age, the study would warrant repetition in age matched samples. While we acknowledge that a small proportion of the Chinese students had a medical or health care background, and that a small proportion of the other students (medical or from ANU) were Chinese, this would most probably reduce the likelihood of finding differences attributable to culture or training. As the instrument has been designed in consultation with professional leaders in a wide range of Asia-Pacific countries, we now encourage utilisation of at least the English version of the instrument in relevant studies among English-literate groups within those countries. A second set of studies will now address the utility of translated versions of the instrument for use in other population groups not literate in English within those countries.

Within the Australia setting, the differences detected between the student groups examined are perhaps unexpected and worthy of further comment. Specific health care training clearly has an association with increasing recognition of the general health burden due to depression. This association also appears to increase with the degree of that training and exposure to direct clinical experience of mental disorders (i.e. fourth versus second year medical students). The importance of medical courses increasing their commitment to this style of teaching in Australia and across the region needs to be emphasised [20,21]. Secondly, those students from ethnic Chinese backgrounds in Australia have markedly different knowledge and attitudes towards depression. This does not simply reflect an unwillingness to discuss such difficulties as the relevant self-report instruments indicate moderately high levels of psychological distress among this cohort. The lack of recognition of depression as a general health problem, the tendency to recognise features more typically conceptualised as anxiety, the greater reliance on pharmacists and alternative practitioners rather than traditional medical services and the expectation of greater discrimination all appear to be consistent with attitudes expressed within other ethnic Chinese groups not residing in Australia [22,23]. Such effects have major implications for current public awareness and health literacy interventions in Australia [4,7]. While we are recording substantial shifts in public attitudes among the broad Australian public [12], we may need to deal more effectively with those specific attitudes held by key ethnic groups or those who have recently arrived in our country.

The IDLS does detect significant differences in understanding of depression among groups from different ethnic backgrounds and between those who differ in terms of prior health training. These results provide preliminary support for the suitability of the IDLS for use in a wider international context.

ANU- The Australian National University;

DALYs- Disability Adjusted Life Years;

GUPI- General-practice Users' Perceived-need Inventory;

IDLS- International Depression Literacy Survey;

SEBoD- "Reducing the Social and Economic Burdens of Depression" in Asia;

SPHERE- Somatic and Psychological HEalth Report.

The author(s) declare that they have no competing interests.

IBH designed the survey and drafted the manuscript. TAD participated in the survey design and drafting the manuscript. GML participated in the survey design, performed the statistical analysis and participated in drafting the manuscript. YR participated in carrying out the study surveys and drafting the manuscript. MLH participated in carrying out and coordination of the study. MIB participated in statistical analysis and drafting the manuscript.

All authors read and approved the final manuscript.

The pre-publication history for this paper can be accessed here:

The limited nerve sources available for the reconstruction and restoration of upper extremity function is the biggest obstacle in the treatment of brachial plexus injury (BPI). Recently, there have been several techniques invented to solve this. One is the use of the contralateral healthy cervical seventh nerve root (CC7) reported by Gu et al. in 1992 [1]. They used this as a donor nerve to reconstruct the median nerve function of the affected arm. Another is the use of a part of the ulnar nerve transfer for reinnervation of the biceps brachii muscle for the treatment of upper type BPI reported by Oberlin et al. in 1994 [2]. Several authors have reported excellent outcomes for nerve recovery with the use of these techniques without leaving functional loss of the donor nerves [3-7].

We describe a 21-year-old patient with BPI who received a vascularized free gracilis muscle transfer [8] innervated by a part of the vascularized ulnar nerve connecting CC7 to the affected median nerve. This successfully restored wrist motion and sensation to the palm.

A 21-year-old man sustained contusion injuries to both lungs and the brain in a motorcycle accident. He was transferred to a hospital, where he was treated for the lung and brain injuries. As his consciousness recovered, his left arm was found to be completely paralyzed. He was diagnosed as having a total left BPI, and underwent surgery for this three months after the injury. During surgery, the left third, fourth and fifth intercostal nerves were transferred to the left musculocutaneous nerve [9], the sixth and seventh intercostal nerves to the left thoracodorsal nerve, and the left accessory nerve to the left suprascapular nerve [10] (Fig. 1).

Twelve months after the BPI surgery performed in the previous hospital, he visited our clinic complaining of complete loss of sensation and motor function distal to the left elbow joint. Reasonable recovery of the shoulder and elbow joints was seen from a physical examination. The left elbow flexion range was 100° against gravity and the abduction and flexion ranges of the left shoulder against gravity were 80° and 85°, respectively. However, the elbow extension was M0 according to the Medical Research Council Scale (MRCS) and the range of left shoulder extension was 15°, which was mostly achieved by motion of the left scapula. No active motion was observed in the left wrist and fingers. The patient demonstrated complete numbness in the area innervated by the left C5-T1 nerve roots. He had slight stiffness in the MP joints of the left index finger to the little finger, and started rehabilitation exercises to soften the joints. No other joints of his left upper extremity showed restriction of the range of motion in the passive movement. Elbow flexion was recovered to M4 level according to MRCS at 15 months after surgery thanks to muscle-strengthening exercises.

Eighteen months after the injury, he underwent reconstructive surgery to his left wrist and hand. As the first step of the reconstruction, the CC7 was transferred to the median nerve through a vascularized ulnar nerve taken from the affected arm. The left ulnar nerve was elevated based on the superior ulnar collateral vessels, including the ulnar vessels distally. A monitor skin flap (2.5 × 1 cm) supplied by perforators arising from the ulnar vessels was also harvested from the distal third of the forearm together with the ulnar nerve. The ulnar nerve and vessels were sectioned at the wrist level distally. The ulnar nerve was also sectioned just proximal to the insertion of the superior ulnar collateral vessels into the ulnar nerve in the upper arm level proximally (Fig. 2). The median nerve was sectioned at the same level of the proximal section of the ulnar nerve. The proximal stump of the transected ulnar nerve segment was approximated to the distal stump of the median nerve. The distal stump of the ulnar nerve segment including the ulnar vessels was transferred to the right neck through a subcutaneous tunnel and approximated the posterosuperior portion of the right seventh cervical nerve root as described by Songcharoen et al. [5] (Fig. 3). The distal stump of the ulnar artery accompanying the transplanted ulnar nerve segment was approximated to the transverse cervical artery in the right neck followed by ligation of the distal stumps of the ulnar veins. After surgery, the monitor flap survived completely. The patient showed transient paresthesia in the tips of right index and middle fingers, which lasted for two months. Grip strength was not affected. The sensory recovery of the affected limb was rapid, and touch sensation was restored to the midpalm level of the affected hand by four months after the CC7 transfer.

As the second stage of the reconstruction, the free vascularized gracilis muscle taken from the right medial thigh was transplanted to the left upper arm four months after the CC7 transfer. The proximal stump of the gracilis was joined to the clavicle with bone anchors and to the trapezius with 1-0 Ethicon sutures. The distal stump of the transplanted muscle was passed under the brachioradialis muscle, stretched maximally and sutured to the extensor carpi radialis brevis tendon in interlacing fashion, maintaining the shoulder joint at 60° abduction and 30° flexion, and the elbow joint at 45° flexion. The tension of the transplanted muscle was set higher than normal, because a slight flexion contracture of the elbow joint would help the patient bend the elbow joint. The medial circumflex femoral vessels of the transplanted gracilis muscle were approximated to the left transverse cervical vessels, as the thoracoacromial vessels were not suitable to microvascular anastomosis because of severe scarring. In the left anterior axillary area, two funiculi of the ulnar nerve (about 10% thickness of the original nerve) were separated from the vascularized ulnar nerve–having been transplanted in the previous operation–and taken proximally for about 1.5 cm. The proximally based partial ulnar nerve stump was approximated to the anterior branch of the obturator nerve innervating the free gracilis muscle (Fig. 4). The stump size of the two funiculi of the ulnar nerve matched that of the branch of the obturator nerve. The distance between the site of the neurorraphy and the entrance of the obturator nerve to the gracilis muscle belly was about 3 cm.

Postoperative recovery was uneventful. After the muscle graft operation, the patient's sensation in the left palm was not compromised and showed continuous further recovery, even though almost 10% of the nerve fibers had been taken to innervate the transplanted muscle. Contraction of the transplanted muscle was observed eight months after the transplantation. Fifteen months after the muscle graft, the dorsiflexion of the left wrist was 35° against gravity with the elbow joint held fully extended (approximately at a 30° -flexion position) when the patient applied forces to the flexor muscles of his right wrist and ulnar fingers (mainly, right and little fingers). His right fist formation and slight palmar wrist flexion was synchronously accompanied by the dorsiflexion of his left wrist (Fig. 5A and 5B). The left elbow joint was restricted in extension after the muscle transplantation; the active range of the joint was -30° in extension and 130° in flexion. The flexion force of the joint was augmented by the muscle transplantation and recovered to an M5 level. Recovery of touch sensation advanced to the base of the index and middle fingers and thumb. The sensation in the left palm was S2 on the MRCS. When the tips of the fingers and thumb were touched or scrubbed, he did not recognized it as a sensation of his left digits but as a weak tingling pain at his right middle fingertip. The sign of motor recovery of the left median nerve has not been observed yet.

As the final stage of the reconstruction, we plan to perform biceps brachii muscle transfer to the triceps muscle to stabilize the elbow joint, tenodesis at the wrist and arthrodesis of the thumb to obtain pinch function. The patient has not yet agreed to final operation, because he is completely satisfied with the improved muscle strength of the elbow flexion and is worried about possible loss of elbow flexion strength after the biceps brachii muscle transfer.

This patient with BPI involving the total brachial plexus had undergone intercostal nerve transfer [9] to the musculocutaneous nerve and a spinal accessory nerve transfer [10] to the suprascapular nerve in a previous hospital. We restored sensation of the palm and motion of the wrist to the patient by transplantation of a vascularized ulnar nerve between the CC7 and the median nerve and a free vascularized gracilis muscle innervated by a part of the transplanted vascularized ulnar nerve. Nineteen months after our first operation, the affected wrist was flexed dorsally by the transplanted muscle innervated by a part of the vascularized nerve connecting to the CC7 and touch sensation recovered up to the base of each fingers.

Several authors have reported excellent outcomes of the sensory and motor recovery of the median nerve using a vascularized nerve transfer connected to the CC7 [1,3,5,7]. Poor motor recovery is to be expected in muscles denervated for more than six months, and skin sensation does not recover if it has been denervated for a long time, because its muscle fibers or sensory organs become too atrophied [11,12]. In our patient, median nerve reconstruction was performed 16 months after the original injury. Excellent recovery of the sensory and motor function of the median nerve could not be expected in this patient, even with CC7 transfer to the median nerve. Thus, it was controversial whether this patient had a valid indication for a CC7 transfer to reconstruct the median nerve. Moreover, for this patient no motor source was available without the vascularized nerve connecting CC7. The intercostal nerves and the accessory nerve had already been used to restore elbow flexion and shoulder mobility. The patient had left phrenic nerve palsy. The patient had no chance of motor recovery without using a free functioning muscle transfer, as no muscles in the affected arm would function even after the neurotization because of the prolonged loss of innervation. Therefore, we decided to perform a CC7 transfer to the median nerve, expecting it to serve as the motor source of a free functioning muscle transfer as well as for recovery of the sensation of the hand. Because a movable hand lacking sensation is no more harmful than a totally paralyzed hand, we reached a consensus with the patient before surgery that the free gracilis muscle transfer would not be done if the sensory recovery of the median nerve was poor. Actually, the patient showed more rapid and better sensory recovery in the median nerve than we had expected before the CC7 operation. Four months after the operation, the patient recognized touch sensation up to the midpalm level of the hand. The sensory recovery of this patient was extremely faster than that reported by previous authors [1,3,5,7]. Although there may be different opinions [7], one possible explanation of the rapid nerve recovery through the CC7 operation in this patient is that the nerve recovery in the transplanted ulnar nerve might have been accelerated by sufficient vascular supply supercharged by the accompanying ulnar artery connected to the transverse cervical artery in the right neck. Most previous authors performed CC7 operations without supercharging the ulnar vessels accompanying the transplanted ulnar nerve [1,3,5]. In addition, the patient was still 22 years old when he had the CC7 operation. His youth might have also contributed to his rapid sensory recovery of his affected median nerve.

Several surgeons have treated patients with upper BPI using a part of the ulnar nerve transferred to the biceps brachii branch of the musculocutaneous nerve and have demonstrated rapid reinnervation of the biceps brachii muscles without leaving any functional loss in the donor ulnar nerve [2,4,6,10]. Oberlin et al. reported that no functional loss should occur in the ulnar nerve after partial ulnar transfer, because the nerve fibers are still mixed and their locations are not distributed functionally at the midarm level [2]. In our patient, no regression of the sensation of the affected hand was observed after a part of the vascularized nerve was transferred to a branch innervating the free transplanted muscle. A part of the transplanted ulnar nerve was taken in the subclavicular region and joined to the obturator nerve of the transplanted gracilis muscle. The ulnar nerve was sectioned just above the insertion of the superior ulnar collateral vessels to the ulnar nerve, and the distal portion of the nerve was reflected proximally in the CC7 operation. Thus, the portion where a part of the transplanted ulnar nerve had been harvested was almost the same as where a part of the ulnar nerve had been taken in Oberlin's original procedure [2]. Because the nerve fibers had not been distributed functionally in the portion where a part of the transplanted ulnar nerve was taken, the palmar sensation in this patient after harvesting a part of the transplanted nerve may not have been compromised.

In conclusion, part of the transplanted vascularized ulnar nerve connected to the contralateral healthy cervical seventh nerve root can be used successfully as a motor source and may be available in the treatment of patients with BPI with scanty motor sources.

The author(s) declare that they have no competing interests.

RK was responsible to this patient for the whole treatment of the brachial plexus injury and performed the surgery as a main surgeon. RI and KN assisted RK in the surgery. TN supervised the surgery and gave several suggestions to complete this manuscript. All authors read and approved the final manuscript.

The Japanese national emergency medical service system has been established since 1963 and enables those in need of urgent medical treatment to summon an ambulance by calling the free national emergency telephone number '119'. Approximately 78% of ambulances are staffed by emergency life-saving technicians, who were introduced in 1991. Prior to that, emergency medical technicians played a major role in prehospital settings but were only allowed to perform basic life support procedures such as external chest compression and ventilation with a bag valve mask. Emergency life-saving technicians, by contrast, are permitted to perform endotracheal intubation, defibrillation, intravenous infusion of Ringer's solution, and administration of epinephrine. However, these treatments are allowed only for cardiopulmonary arrest patients. In addition, they are not permitted to carry out additional life-saving interventions and clinical tests such as needle thoracostomy, cricothyroidotomy, 12-lead electrocardiograms, blood glucose measurements and administration of drugs other than epinephrine. Furthermore, the frequency with which vital signs are taken at the scene is still low in Japan; 25% for blood pressure and 27% for SpO₂ [1]. As a result, prehospital care in Japan is very limited compared with to in western countries [2-5].

Meanwhile, the Japanese national medical insurance system and free ambulance call-outs have resulted in a lack of concern over ambulance usage. This usage has increased consistently by over 5 million cases per year since 2004. According to national data, 51% of these were mild cases that did not require hospitalization [1]. In addition, the rapid aging of society, with more than 19% of the population aged over 65 years in 2004 [6], has promoted an increase in ambulance usage, especially for acute disease [7,8].

In this situation, refusal of ambulance transfer for mild cases is being debated nationally. The fire and disaster management agency established a 'Committee for demand of ambulance usage' in 2006. For appropriate ambulance usage for hospitalization, a simple triage tool to decide the refusal of ambulance usage would be useful.

The purpose of this study was to investigate the relationship between prehospital data and hospitalization outcome for acute disease patients aged ≥ 15 years, and to develop a simple tool for deciding the refusal of ambulance usage based primarily on vital signs.

This study is an investigation of the relationship between prehospital data and requirement for hospitalization in acute disease patients aged ≥ 15 years; it also details the development of a simple triage assessment tool. Age, systolic blood pressure, pulse rate, level of consciousness, SpO₂level and ability to walk were found to be statistically significant predictors for hospitalization requirement. A prehospital score ranging from 0 to 14 was developed on the basis of the multivariable analysis results.

In western countries, prehospital triage guidelines are based on patient symptoms for acute disease [10-16], but these are inappropriate for use in Japan as emergency life-saving technicians can only perform limited medical treatments.

There has been only one prehospital triage tool for acute disease and trauma in Japan: the prehospital severity and urgency criterion, produced by the Japan Foundation for Ambulance Service Development in 2004 [17]. However, this criterion is complex to use because it is based on 10 categories: trauma, burn, intoxication, consciousness disorder, chest pain, dyspnea, gastrointestinal bleeding, abdominal pain, pregnancy, and infancy; it requires approximately 20 items to be checked. In practice, this criterion is not used because of its complexity. Moreover, criteria based on chief complaints are of little use for cases with complaints outside those 10 categories, such as paralysis or headache [11].

In contrast, the prehospital score of the present study is a simple to use, comprehensive triage tool for acute disease. Furthermore, emergency personnel without emergency life-saving technician licences could use this scoring system because of its simplicity.

Sensitivity, specificity, PPV and NPV are barometers for the utility of a triage tool. As under-triage is less desirable than over-triage, sensitivity is more important than specificity. When we set the cut-off point at total score ≥ 2, sensitivity and specificity were respectively 97% (2,900/3,002) and 16% (841/5,328). Furthermore, we can also predict the likelihood of death as 0% and 100% when the score is low (≤ 1) or high (≥ 12), respectively.

If ambulance transfer is refused for patients with score ≤ 1, 16% (841/5,328) of cases of inappropriate ambulance usage would be avoided. On the other hand, this decision made 3.5% (102/3,002) into under-triage cases.

Previous studies in western countries have reported prehospital guidelines for predicting the requirement for admission to the emergency department (sensitivity = 90% and specificity = 37%) [11] and a prehospital protocol for predicting the critical event during transport (sensitivity = 95% and specificity = 33%) [12]. These studies relied on detailed protocols according to symptoms. Although simple comparison with our study is difficult because of differences in the medical systems, our scoring system showed higher sensitivity and lower specificity. To avoid under-triage, lower specificity might be inevitable.

However, this study is not without its limitations. First, the scoring system did not include information other than the six fundamental elements. For example, the chief complaint is often strongly associated with outcome: chest pain, paralysis, hematemesis or melena. Patients with gastrointestinal bleeding usually require hospitalization, yet 10 patients in this study with gastrointestinal bleeding scored ≤ 1. Moreover, 4 cerebral infarction cases, 1 myocardial infarction case, 1 pneumothorax case and 1 tuberculosis case scored ≤ 1. These cases needed rapid transfer to an emergency hospital. If low-scoring (<=1) patients with chest pain, paralysis, hematemesis or melena were considered as indicating ambulance usage, 17% (17/102) of under-triage cases could be avoided. Since our purpose is to develop a simple triage tool that would be useful for ambulance refusal for Japan emergency personnel regardless of emergency life-saving technician licence, an additional assessment for chief complaints would be necessary for practical use. However, sensitivity and specificity did not change substantially when these chief complaints were taken into account (97% and 15%, respectively).

Second, the scoring system was mainly based on patient vital signs, but did not include body temperature or respiration rate. The method for measuring body temperature in the prehospital setting is not standardized; some are measured by tympanic temperature, others by axillary temperature. Therefore, we did not use body temperature. The inclusion of respiration rate in the regression model excluded 15% of the subject data and did not improve the predictive value for hospitalization (sensitivity = 97% and specificity = 15%).

Third, the proportion of emergency hospitalization in Kishiwada City is substantially lower than the national average: 36% compared with 49%. Criteria for hospitalization vary according to areas and hospitals. Furthermore, we could not obtain complete data for 9% of patients in Kishiwada City. Therefore, the external validity of the prehospital score is uncertain.

Fourth, we had no follow-up data on patients who were not hospitalized. Some might be misdiagnosed in the emergency room and not hospitalized at that time, but would then be hospitalized after a few days.

Despite these limitations, the present study suggests that a simple score based on 6 fundamental elements enables us to decide whether ambulance transfer is indicated with 97% sensitivity and 16% specificity. In the near future, refusal of ambulance transfer for mild patients may be allowed in Japan. Our prehospital score could be a tool for deciding transfer refusal by emergency personnel.

We have developed a prehospital score, a simple acute disease triage tool for Japan's emergency medical service system. Further research on validity would be necessary.

The author(s) declare that they have no competing interests.

YM, HT and HF were involved in the acquisition of data. YT was involved in data analysis and drafting the manuscript. TT and HI provided statistical advice on study design and critical revision of the manuscript. All authors read and approved the final manuscript

The pre-publication history for this paper can be accessed here:

Diabetes is a chronic illness which requires the individual to assume responsibility for their own care with the aim of maintaining glucose and blood pressure levels as close to normal as possible [1]. Maintaining optimal glucose and blood pressure levels reduces the risk of diabetes related complications [2,3]. Treatment of diabetes involves psychological, social and physical adjustments to an individual's lifestyle [1]. This can be confusing and overwhelming for people with diabetes [4]. They have to make a complex range of lifestyle modifications sometimes without necessarily noticing any tangible effects [4]. Emotional and quality of life issues need to be attended to as well as physical issues [4].

Diabetes self management training has traditionally been delivered in a didactic setting with emphasis on imparting knowledge. However this approach has been shown to be ineffective in individual behaviour change and improving metabolic control [5]. In recent times alternatives to the traditional delivery of diabetes care have been investigated, for example, the concept of peer support which emphasises patient rather than professional domination [1]. Peer support could be implemented to complement existing diabetes care. Structured care for people with type 2 diabetes in general practice is not yet well established in the Republic of Ireland [6]. Prevalence of diabetes amongst people over 40 years of age attending 41 general practices in the Republic of Ireland reported a prevalence of type 2 diabetes of 9.2% indicating similar prevalence figures to other European countries [6]. The usual care of patients with type 2 diabetes in the Republic of Ireland is outlined in Figure 1.

Testing a complex intervention such as peer support presents a challenge to researchers. Complex health interventions are built up of several components which may include organisational and delivery methods [7,8]. The fact that they involve a number of separate components presents difficulties in isolating the "active ingredient" of the intervention that is effective [8,9]. Therefore it is recommended that a complex intervention for an RCT should be carefully planned and designed [10]. To guide researchers, the UK Medical Research Council (MRC) devised a five phase framework for developing and evaluating RCT's of complex interventions [8]. The framework is comprised of five phases [8]. The Pre-clinical phase involves establishing a theoretical basis to support the intervention. Phase 1, modelling, involves developing an understanding of the intervention and its possible effects. At this point the components of the intervention are delineated. These first two phases are often interrelated. Phase 2, the exploratory trial, is crucial. This is a test of the feasibility of key components of the intervention. Phase 3 is the definitive RCT. Finally long term implementation of the intervention is examined in Phase 4. A flowchart of the methodology of the application of the framework is presented in Figure 2.

This framework has been utilised in a variety of RCT's that have evaluated complex interventions in primary care [7,11-13]. These RCT's examine professionally led interventions for example a behaviour change intervention delivered by primary care practitioners to patients with coronary heart disease [10]. This paper is the first to examine the development of a complex intervention involving peer support.

We describe below, the application of the first three phases of the MRC framework which led to the development of the intervention of peer support in type 2 diabetes based in primary care. The definitive intervention is currently being tested in a cluster RCT, the peer support in diabetes study (Table 1).

Ethical approval has been obtained from the Ethics Committee of the Irish College of General Practitioners (Protocol No.: REC0904-11; 01/12/04)

The MRC framework was instrumental in the development of a robust intervention of peer support in type 2 diabetes in primary care. The intervention of peer support was considered in depth incorporating an analysis of current literature, qualitative work with those who would be both experiencing, delivering and administering the peer support system and finally an analysis of how the intervention would run in the pilot study. It enabled a clear and detailed understanding of the components of the intervention and how each should be documented and tested during the definitive study. The effectiveness of this intervention is now being tested in a cluster randomised controlled trial involving twenty general practices and 420 patients with type 2 diabetes.

The author(s) declare that they have no competing interests.

All authors reviewed and approved the final version of this manuscript. GP, SMS, TOD and DW designed the study, prepared the protocol and participated in writing this paper. GP conceptualized and drafted the paper, and conducted the data collection. FOK prepared the protocol and participated in writing the paper.

The pre-publication history for this paper can be accessed here:

During the last decades in Mexico, as elsewhere, the population has aged, causing an increase in the level of chronic degenerative diseases and a consequent increment in medication. Polypharmacy is now common, and carries a high risk of drug-drug interactions and drug-disease interactions. These may cause adverse effects, or the therapeutic effects of the combined medicines may change, with serious consequences for health. In the United States 25% of ambulatory patients taking drug combinations were at risk for clinically important interactions [1]. Furthermore, it has been reported that about 40% of hospitalized patients had at least one potential drug-disease interaction [2]; a large study including 70,203 outpatient visits by patients aged 65 and older found that 2.6% ofvisits with at least one prescription had one or more of the 50 inappropriate drug-disease combinations examined [3]. A European study of 1601 ambulatory elderly patients, taking an average of seven different drugs, found that 46.0% were at risk for at least one clinically important potential drug-drug interaction [4]. It has been shown that inappropriate prescription combinations increase with patient age, are more frequent in men, rise in line with the Charlson co-morbidity index [5], and increase as the number of prescribed drugs increases [2,3,6-8]. It is possible that other risk factors for potential interactions exist, and these should be identified to establish successful methods for improving prescription practices.

Adverse consequences of drug interactions have been shown in various studies. Drug-drug interactions cause 4.8% of hospitalizations attributed to drugs in the elderly [9-11]. In most cases they are erroneously interpreted as patient deterioration because of illness, poor adherence to prescribed treatment, or infection [12].

In Mexico, inappropriate prescription of drugs with potential interactions causing serious risks to patient health has been little studied. Work in this area has focused mainly on hospitalized patients, with only specific drug combinations analyzed [13]; moreover, the studies have not produced conclusive results [14,15].

In the present study, we sought to determine the frequency of potential drug-drug and drug-disease interactions in prescriptions of ambulatory patients over 50 years of age, who attended to Mexican Institute of Social Security (IMSS) family medicine clinics. In addition, we aimed at identifying factors associated to them.

The present study is a secondary data analysis from the Educational Strategy Study (ESS) involving both doctors and patients over 50 years of age, focused on improving utilization of non-opioid analgesics in primary care study and carried out during 2006 in two large IMSS family medicine clinics (FMC) located in Mexico City.

IMSS is the largest medical institution in Mexico that provides health care for more than 53 million of Mexicans. The provision of services is divided into three levels, where the Family Medicine Clinic is the primary care level. IMSS FMC range from 2 to 40 examining rooms; the clinics included in the study had 32 examining rooms, which means 64 family doctors working in the morning and evening shifts (32 in each);. These clinics were selected by convenience and are similar in organization to the rest of those that constitute IMSS primary care system. Cross-sectional data from 624 ambulatory patients over 50 years of age who made visits to 127 family doctors (96% of all family doctors) were collected. On average, five patients seen by each family doctor were included consecutively. The patients studied suffered from non-malignant pain syndrome and received prescriptions of non-opioid analgesics for 7 days or more. The variables analyzed were general characteristics of the patient (age, marital status, literacy), medical history (number of chronic diseases, number of medicines prescribed), and complete information about all oral and injected drugs prescribed by the doctors during the consultation (including occasional drugs and those that were prescribed for regular use, besides the prescription of non-opioid analgesics described as inclusion criteria). All patients were personally interviewed immediately after the visit and the information from the electronic medical record and from the electronic prescription was registered. Most of information (diagnosis and prescriptions) was obtained through personal interview and additional data were obtained from the electronic medical charts and the electronic prescriptions.

The International Classification of Disease, version 10 (ICD-10) [16], and the Anatomical Therapeutic Chemical classification system [17] served to codify the data. To look for potential interactions every combination of prescribed drugs was analyzed by using the Thompson Micromedex program [18]. Drug-drug interactions were sorted by clinical relevance using the Classification System of the Department of Pharmacology, Hospital Huddinge, Stockholm, Sweden [19]. In this classification, drug interactions are rated A and B when they are not of clinical importance (type A), or the effect of the interaction has not yet been established (type B). One interaction of type C can cause possible changes in the therapeutic effects, or may cause adverse effects, but can be avoided adjusting the individual drug doses. A potential drug-drug interaction of type D indicates a potential for severe adverse effects; individual dose adjustment is difficult in these cases. For this paper, only drug-drug interactions of type C and D were detected and analyzed. Using the Zhan Classification all drug-disease interactions were classified as being of low, moderate or high clinical significance [3].

The IMSS Ethics Committee approved the ESS (registration number: 2005-785-185).

The median age of patients was 69 years (range 50–94 years). Most (78.7%) were women, 63.9% were housewives and more than half were single, divorced or widowed. The average number of chronic diseases per patient was 3.4 ± 1.5 and the most frequent illnesses were degenerative joint disease, hypertension, chronic gastritis, diabetes mellitus and dyslipidemia (Table 1). The total number of medicines prescribed to the 624 patients was 3,739 with an average of 5.9 ± 2.5 drugs per patient. The most frequent drugs prescribed were active on the alimentary tract, or affected general metabolism (drugs combating gastrointestinal diseases were the most commonly prescribed drugs). The next most common class of drugs was active on the cardiovascular system; drugs addressing muscle-skeletal system problems were next in prescription frequency, and, finally, drugs active on the nervous system. In this final group paracetamol was the main drug consumed (Table 2).

About 80% of patients had prescriptions for one or more combinations in the potential drug-drug interaction class. The most frequent drug interactions were type C, such as combinations of non-steroidal anti-inflammatory drugs (NSAIDs) with antihypertensive drugs (40.4%), and with low doses of acetylsalicylic acid (ASA) (34.0%). 3.8% of patients were prescribed drug combinations with interactions that should be avoided, and two patients were prescribed drugs with two potential type D interactions. Also, 400 patients (64.0%) had one or more potential drug-disease interactions of moderate clinical significance, given that the medicines prescribed could cause either light or moderate adverse effects. The most frequent were NSAIDs in patients with hypertension and/or chronic heart failure; β-blocking agents in patients with diabetes mellitus, and NSAIDs in patients with chronic renal failure (Table 3) When analyzing both classes of potential interactions we did not find statistically significant differences in patient gender, marital status or education.

Table 4 shows bivariate analysis of the relationship between the characteristics of patient and prescription, and potential drug-drug interactions. The variables found significant (p < 0.05) were, patient older than sixty years, 3 or more diseases, cardiovascular disease, endocrine, alimentary and metabolic disease and receiving five or more medicines.

The factors significantly associated with having one or more potential drug-drug interactions in the logistic regression model included: taking five or more medicines (adjusted odds ratio (aOR): 4.34, 95%CI: 2.76–6.83), patient age of 60 years or older (aOR: 1.66, 95% CI: 1.01–2.74) and suffering from cardiovascular diseases (aOR: 7.26, 95% CI: 4.61–11.44) (Table 5); the adjustment variables were: number of disease ≥ 3 and endocrine, alimentary and metabolic disease.

The bivariate analysis of the factors associated with drug-disease interaction did not show any statistically significant association; therefore, a logistic regression analysis was not performed for these interactions.

Various studies have shown that potential drug-drug and drug disease interactions are frequent when patients receive multiple prescriptions. This is true for both ambulatory and hospitalized patients, and, in many cases, causes adverse effects and changes in therapeutic efficacies of the combined medicines, with consequent poor control of the diseases under treatment [1-4,6-12].

In the present study, we found that the frequency of potential drug-drug interactions in prescriptions of family doctors working in primary care clinics in Mexico City was almost 80.0%; this is higher than the frequency in Europe [4] (46.0%) and in the United States [1] (25.0%) in ambulatory patients over 59 years of age. The rates we found may be unique for the sample and may not be fully representative of Mexican population situation. The higher prevalence of potential drug-drug and drug-disease interactions in this study compared to others studies is likely attributable to the characteristics of the study sample (older adults with very high prevalence – nearly 90.0% – of NSAID utilization) among other reasons.

The frequency of type D interactions (which should be avoided) was smaller (3.8%) in our work, when compared with other studies (the type D frequency was 10.0% in the European study) [4]. The three combinations with drug-drug potential interactions that were found most frequently in our work are among those reported by Bjorkman [4]. These drug interactions are of type C and, without dose adjustment and patient monitoring, such interactions may antagonize drug effects on vascular tone and may result in increases in blood pressure (angiotensin-converting enzyme (ACE) inhibitors + NSAIDs, ACE inhibitors + low doses of ASA) [20] or may increase gastrointestinal adverse effects (NSAIDs + low doses of ASA) [21]. These finding coincide with the literature review performed by Becker et al., in which it was reported that drugs most often responsible for hospital admissions were NSAIDs and cardiovascular drugs, and the most common causes for such admissions were gastrointestinal tract bleeding and hyper- or hypotension [11].

Other authors have reported that both types C and D show similar hospitalization frequencies [10]. Therefore, medical doctors must be familiar with both interaction types.

We found that not only potential drug-drug, but also drug-disease interactions, were frequent in prescriptions. The frequency of the latter (64.0%) found in the present work is greater than reported in either hospitalized patients [2] or ambulatory patients [3].

The most frequent were interactions involving NSAIDs that were prescribed to patients with hypertension and/or chronic heart failure, and prescriptions of NSAIDs and ACE inhibitors. Although this finding was influenced by patient inclusion criteria, it shows that the flaws in the knowledge of prescribers regarding interactions of this group of drugs. In such cases, other therapeutic options should be considered to avoid potential drug interactions, like using paracetamol to manage osteoarthritis in patients with hypertension being treated with ACE inhibitors.

In general, our findings agree with other studies reporting that the risk of potential drug-drug interactions increases with each new prescription issued and with the aging of the patient [2,3,6-8]. In line with this, we found that patient age of 60 years or older and taking 5 or more medicines increases the risk of such potential interactions. Polypharmacy is an important problem in older people that has been reported as a frequent event all over the world [3,4,22,23]. In our sample, the patients took an average of 5.9 drugs, and those 60 years or older took an average of 6.1 drugs (37.3% of them took 7 or more medicines).

Within the context of this study, there exists limited published local information regarding the average number of drugs that a patient older than fifty years gets prescribed. Previous studies have been carried out in specific groups of older patients, such as community-dwelling elderly hospitalized due to inappropriate drug prescriptions [24], in nursing home residents [25] and in patients with hypertension [26]. For example, in hospitalized elderly patients, the average number of drugs consumed was 6.0 [24], and in nursing home residents was 2.8 [25], and in the study addressing patients with hypertension [26] aged 60 years and older, only 1.9% were taking three or more hypertensive drugs; yet, in this study the consumption of other drugs was not analyzed. Further studies should be advisable to gain in depth knowledge about the average number of drugs that family doctors prescribe to the elderly in Mexico.

Furthermore, we found that having a cardiovascular disease increases seven-fold the risk of potential drug-disease interactions. This means, as has been recommended in prior studies from other countries, that doctors need to pay more attention to drug prescription and patient monitoring when treating older individuals [2,3,6,7] also, the patients with cardiovascular diseases deserve more attention. It is necessary to consider interventions to reduce the drug interaction problem. Computer-based access to information on all prescriptions dispensed, and automated doctor alerts on the most frequent potential drug interactions encountered, would be most helpful. These tools are effective in reducing inappropriate prescriptions, and doctor acceptance in other populations has been reported [27,28]. Alternatives such as programs of continuous medical education, or pharmaceutical support, may also be considered. It has been found, however, that even pharmacists cannot detect all potential drug interactions because their number rises dramatically as the number of medicines prescribed increases [29].

Among the limitations of this study is that it only permits an approximation to problem of drug interactions in family medicine practice. The patients' group studied was very limited (all patients had non-malignant pain syndrome). We believe that some drug interactions may be more frequent in such patients. For example, we found that the most frequent potential drug-disease interaction involved the use of NSAIDs in patients with hypertension. In patients with acute or infectious disease is possible that other interactions would be found more frequently. Also, in this cross-sectional study we did not determine any possible relationship between drug-drug nor drug-disease interactions and the health status of the population studied.

It is possible to conclude that the high frequency of prescription of drugs with potential drug interactions is common in primary care level; the easiest way to reduce the frequency of them is to decrease the number of medicines prescribed. Nevertheless, sometimes it is difficult to reduce the number of drugs prescribed for patients with multiple chronic conditions; therefore, to lower the frequency of potential interactions it could be necessary to make a careful selection of therapeutic alternatives, and in cases without other options, patients should be continuously monitored to identify adverse events.

The author(s) declare that they have no competing interests.

SVD, HRM contributed in the conception and design of the study, literature review and statistical analysis. LPTA reviewed for important intellectual content. MSO contributed in data management. All authors participated in the interpretation of data and read and approved the final version to be published.

The pre-publication history for this paper can be accessed here:

Strategically located on the alveolar surface, alveolar macrophages represent highly specialized macrophages that function primarily in lung defence against inhaled particle matter, microorganisms and environmental toxins. Among microorganisms, gram-negative bacteria and more precisely, the lipopolysaccharide (LPS) component of the outer cell wall, is a very potent activator of macrophages. LPS binds to LPS-binding protein and is delivered to the cell surface receptor CD14, before being transferred to the transmembrane signaling receptor toll-like receptor 4 (TLR4) and its accessory protein MD2 [1]. LPS stimulation activates several intracellular signaling pathways including the three mitogen-activated protein kinase (MAPK) pathways: extracellular signal-regulated kinases (ERK) 1 and 2, c-Jun N-terminal kinase (JNK) and p38. These signalling pathways in turn activate a variety of transcription factors which coordinate the induction of many genes encoding inflammatory mediators as well as anti-inflammatory cytokines.

The control of inflammatory responses is critical to the host to allow resolution and avoid tissue damage. IL-10 is a key anti-inflammatory factor and pleiotropic cytokine produced by a variety of cell types among which monocytes/macrophages are the main sources [2]. IL-10 mediates the inhibition of pro-inflammatory cytokines such as TNF-α, IL-8, IL-6, IL-1β, IL-12 [3-7]. IL-10 has also been shown to inhibit antigen-presenting cell function, including the maturation of dendritic cells [8] and the expression of MHC class II and co-stimulatory molecules [9,10]. IL-10 gene regulation can occur both at the transcriptional and posttranscriptional levels [11]. Several studies have shown that the transcription factor Sp1 plays an important role in IL-10 transcription (an Sp1 responsive element in the IL-10 promoter is localized at -89 to -78) [12-14]. Moreover, detailed studies showed that p38 mitogen-activated protein regulates LPS-induced activation of Sp1 in THP-1, a human monocytic cell line [14]. The STAT3 transcription factor may also bind to an element in the IL-10 promoter gene and the use of a dominant negative form of STAT3 was able to decrease IL-10 transcription [15]. More recently, the protooncogene c-Maf has been shown to be an essential transcription factor for IL-10 gene expression in macrophages [16] while a role for C/EBP in cooperation with Sp1 has also been suggested [17]. However, the intracellular signalling pathways governing IL-10 gene regulation in human alveolar macrophages are poorly understood. Thus, alveolar macrophages are the main source of IL-10 in the alveoli where they play an important role to control lung homeostasis. One study on human alveolar macrophages [18] showed that activation of PKC decreases IL-10 production whereas activation of protein phosphatases PP1 and PP2A enhance IL-10 secretion. In the present work, we evaluate the ability of human alveolar macrophages to produce IL-10 upon LPS stimulation and the role of MAPkinases (ERK, p38 and JNK) and Sp1 transcription factor as intracellular signals leading to IL-10 expression.

Lung homeostasis relies on the equilibrium between the induction of efficient innate defensive responses to inhaled infectious microorganisms and equally effective mechanisms to downregulate the inflammatory response to initiate resolution and tissue repair. As a predominant immune effector cell in the airspaces, the alveolar macrophage is critical to these defence processes. Thus they produce a vast array of cytokines and inflammatory mediators in particular after LPS stimulation, including TNF-α and IL-10 as prototypic pro-inflammatory and anti-inflammatory cytokines, respectively. While the signaling events that mediate TNF-α in HAM have been extensively studied [20-23,19], those responsible for IL-10 production have not been well characterized.

The present study is the first one showing that ERK, p38, JNK and Sp1 are involved and essential in LPS-induced IL-10 expression in HAM. These factors act at the transcriptional level, as IL-10 mRNA stability was not affected by MAPK inhibitors. It is well known that LPS drives intracellular signaling pathways such as MAPKs and NF-κB [23,24] to activate several pro-inflammatory genes including cyclooxygenase-2 [25], inducible nitric oxygen synthase [26], TNF-α and IL-1β [27]. In the present study, we found that in HAM the MAPK signaling pathways were also involved in LPS-induced gene activation of IL-10, a major anti-inflammatory factor. A recent study in murine macrophage raw 264.7 cells also reported that MAPKs were necessary in IL-10 expression by LPS [28] whereas other studies showed that only MAPK p38 was essential for IL-10 expression induced by LPS and other ligands [29-33]. Our study not only showed that all MAPKs were necessary but that they also played an important role in the downstream activation of Sp1 trancription factor.

Sp1 is the founding member of a family of zinc finger transcription factors, which includes at least four Sp transcription factors [34,35]. Among the Sp transcription factors, Sp1 has been extensively studied [36] and is known to be widely expressed and to play a role in the regulation of a vast array of genes. Thus, while not excluding a role for other nuclear factors our data confirm previous studies showing that IL-10 gene expression is controlled by the transcription factor Sp1 [13,14,37] since mithramycin, an inhibitor of Sp1, almost completely abrogated LPS-induced IL-10 production at the mRNA and protein level. EMSA assay confirmed this observation as it showed that mithramycin decreased the activation of the nuclear protein Sp1 after LPS stimulation. Transfection of HAM by antisense oligonucleotides to Sp1 could not be used as another approach to confirm the role of Sp1 in IL-10 induction by LPS, as after the time period required to silence Sp1 expression (minimum 16–20 hrs) following 4 hrs-transfection, HAMs were not anymore responsive to LPS for IL-10 production (data not shown). Interestingly, inhibition of MAPKs by their specific inhibitors, also abolished LPS-induced Sp1 activation. These results are in accordance with previous studies [14,38-42] showing that Sp1 phosphorylation can be induced by ERK and p38 MAP kinases. In addition the present study also showed that JNK MAP kinase is also required for the activation of Sp1 induced by LPS. The three MAP kinases seem however to have different contributions to LPS-induced IL-10 in HAM, with a prominent role of p38 and ERK.

IL-10 production by alveolar macrophages has been debated since some authors described the inability of alveolar macrophages to produce IL-10 [43-45]. Other investigators related this to a reduced production of IL-10 to allergic inflammation [46]. Our data clearly confirm IL-10 production by normal HAM [47-54], provide new information on the mechanisms involved in this production and complete the studies of Boehringer et al [18] who has studied the role of PP1 and PP2A in the regulation of LPS-induced IL-10 in HAM.

Our study demonstrates the contribution of MAP kinases to IL-10 expression in HAM upon endotoxin activation, indicating that ERK and p38 and to a lesser extent JNK are involved. In addition we show that ERK, p38 and JNK are able to trigger the phosphorylation of Sp1, the major transcription factor for the IL-10 gene. These findings are highly relevant to lung immunity, alveolar macrophages assuming front-line defense mechanisms and IL-10 representing a key factor for mucosal tolerance and resolution of inflammatory responses.

The author(s) declare that they have no competing interests.

HC supervised all the experiments, performed the data analysis and wrote the manuscript. AG carried out the experiments. CP was involved in the design of the study and in writing of the manuscript. YS was involved in the design of the study, writing of the manuscript and interpretation of the data. All authors read and approved the final manuscript.