Grupo 3

ORIGINAL PAPER

Effect of a homeopathic-complex medicine on

state and trait anxiety and sleep disorders:

a retrospective observational study

1_{Department of Dietary Nutrition and Complementary Medicine, Azienda Ospedaliera G. Rummo, Benevento, Italy} 2

Latis S.r.l., Genova, Italy

Purpose: The primary objective of this open-label, retrospective, observational study was to examine change in indicators of anxiety changed after treatment period of at least 1 month with the homeopathic-complex medicine Datif-PCÒ. A secondary objective was to describe the quantitative changes in the sleep pattern of the subjects included in the study.

Methods: The study population consisted of 71 subjects suffering from mild to moder-ate anxiety, who were tremoder-ated with a homeopathic-complex medicine. The subjects had attended at least 2 consultations during the year 2010, with the second visit taking place at least 30 days after the first baseline visit. At both visits, subjects were administered the State-Trait Anxiety Inventory (STAI)-Y questionnaire (consisting of trait and state subscales) to assess their level of anxiety. Sleep quality was assessed based on the sub-jects’ self-reported number of hours slept and number of night-time awakenings, before and after treatment.

Results: Compared to baseline, the average (STAI)-state score at visit 2 decreased by more than 6 points, while the STAI-trait score decreased by more than 3 points. This was statistically significant (p < 0.0001) for both subscales. These findings were stable in subgroup analyses (gender, age and anxiety level). The average number of hours slept per night increased significantly at visit 2 compared to baseline (from 5.1 ± 1.1 to 5.56 ± 0.9 h, p < 0.0001). A marked decrease in the average number of night-time awak-enings was also observed at the second visit ( 42%, p < 0.0001).

Conclusions: This retrospective observational study suggests that treatment of anxi-ety and sleep disorders with Datif-PCÒ can produce notable improvements even in a short period of time. Furthermore, the observed effects were not affected by sex, age or baseline anxiety. Further controlled, randomized studies are justified. Homeopathy (2013) 102, 254e261.

Keywords: Anxiety; Retrospective observational study; Low dose;Datif-PCÒ; STAI-Y; Sleep disorders

Introduction

Many sources of everyday stress (such as work, eco-nomic status, or family dynamics) can cause people to feel insecure, worried or anxious, sometimes to the point of interfering with their ability to function. Anxiety is an emotional state so deeply-rooted and widely prevalent among humans that it can sometimes be considered not just a symptom or syndrome, but a way of life marked by escalating feelings of apprehension and/or fear.1Anxiety

*Correspondence: Luigi Coppola, Department of Dietary Nutri-tion and Complementary Medicine, Azienda Ospedaliera G. Rummo, Via dell’Angelo 1, 82100 Benevento, Italy.

E-mail:gico61b@gmail.com

Received 1 October 2012; revised 9 July 2013; accepted 16 July 2013

is defined as a state of apprehension, tension and distress induced by anticipation of a future threat or negative event, accompanied by feelings of dysphoria or physical symp-toms of stress.1e3 The common cause of many forms of anxiety is the subjective perception that one’s skills and personal strengths are inadequate for the magnitude of problems one must confront. Stress and its manifestations (anxiety and, particularly, insomnia) are becoming increasingly prevalent in both clinical practice and everyday life.4,5It is therefore important to develop tools able to detect and measure anxiety, both for psycho-diagnostic purposes and for evaluating the effectiveness of psychotherapeutic interventions. It is important that treatments are not only effective, but also safe and well tolerated.

Insomnia is recognized to be often related to anxiety dis-orders.4e6Sleep can be altered by a delay in falling asleep, or by frequent nocturnal awakenings or early awakenings. Regardless of its cause, a lack of sleep or disrupted sleep will likely lead to poor concentration during daily activities. Thus far, the field of sleep medicine has made remarkable progress, and can now help people who sleep badly by identifying any organic causes, such as breathing problems or endocrine disorders, which when resolved may reduce symptoms of insomnia. The most widely prescribed drugs for the treatment of insomnia and anxiety disorders are benzodiazepines, which have been used for over 50 years.7,8 This class of drugs has recognized side effects such as drowsiness/sedation, disorientation, hallucinations and the possibility of causing eventual long-term drug dependence.8,9A recent, large-scale observational study conducted in the US showed that patients receiving hypnotic prescriptions were associated with a more than 3-fold increase in risk of death.10

Surveys indicate that people suffering from anxiety often turn to homeopathic medicines.11Anxiety is one of the most commonly reported symptoms among French pa-tients who consult homeopathic general practitioners.12 One study found that 50% of Norwegian physicians take a favorable view of homeopathic treatment for certain health conditions, among which anxiety is frequently mentioned.13A study on the use of homeopathic medicines for psychiatric symptoms in the general population found that homeopathy is mainly employed to reduce anxiety symptoms.14

The homeopathic-complex medicine Datif-PCÒis one of the few commercially available options for treating anxiety and emotional disorders. The homeopathic-complex medicine Datif-PCÒconsists of 6 active ingredi-ents diluted and dynamized to 6cH (10 12of the starting substance). These ingredients are traditionally used for anxiety disorders, as reported by the homeopathic refer-ence literature. Datif-PCÒin produced in a tablet formu-lation and this can be an advantage in terms of treatment compliance.

The aim of the present study was to retrospectively observe how indicators of anxiety and sleep quality changed following homeopathic treatment with

Datif-PCÒfor a period of at least 1 month. The primary objective was to evaluate the temporal change scores for the 2 State-Trait Anxiety Inventory (STAI) subscales (STAI-trait and STAI-state).15,16 A secondary objective was to describe the quantitative changes in the sleep patterns of subjects included in the cohort.

Materials and methods

Study population

The study population was drawn from subjects who had been referred to our clinic (Department of Dietary Nutri-tion and Complementary Medicine, at the “Azienda Ospe-daliera Gaetano Rummo”, Benevento, Italy) and who were prescribed homeopathic treatment for anxiety. As a stan-dard procedure for patients who complain of symptoms of anxiety, the homeopathic treatment was prescribed by the homeopath during a short consultation and data con-cerning anxiety (including the STAI-Y questionnaire), sleep disorders and concomitant treatments were recorded on a detailed form at each clinic visit. To select the subjects for our retrospective observational study, we reviewed those forms and recorded which patients met the inclu-sion/exclusion criteria on a study-specific paper case report form (CRF). The inclusion criteria were as follows: 1) age 18e90 years

2) presence of symptoms indicative of anxiety disorder during a first-contact interview

3) diagnosis of mild to moderate anxiety according to the STAI-Y questionnaire (score$20)

4) need for a new anti-anxiety medication or for a treatment to replace their previous medication (sedative-hypnotic drug) because of side effects or because the patient refused to take his/her current medication

5) having attended at least 2 visits between January 1 and December 31, 2010, with the second visit taking place at least 30 days after the first baseline visit.

The exclusion criteria were:

1) presence of other psychiatric disorders

2) inability to comply with the administration of a homeo-pathic medicine

The study was approved by the Independent Ethical Committee of the hospital “Azienda Ospedaliera Gaetano Rummo”, Benevento, Italy, on January 24th, 2011. All the subjects gave their informed consent to the use of their data in this study.

Treatment and concomitant medication: The patients were prescribed the homeopathic-complex medicine Da-tif-PCÒ(Laboratoires Boiron S.r.l.) which is indicated for the treatment of anxiety states associated with sleep disor-ders. The components of Datif-PCÒare: Abrus precatorius 6CH (indicated for spasms, cramps and palpitations), Aco-nitum napellus 6cH (anxiety, tachycardia, oppression), Belladonna 6cH (hyperalgesic spasms, tachycardia, sen-sory hyperesthesia), Calendula officinalis 6cH (spasms), Chelidonium majus 6cH (digestive spasms), Viburnum opulus 6cH (malaise, joint pain, headache).

Subjects were instructed to take 2 tablets 3 times a day for the entire study period (i.e., the time between the 2 visits). The patients purchased the treatment and took it on their own account. No assessment of treatment compli-ance was made. At both visits, patients were asked if they had taken any concomitant medication before the first visit and/or between the 2 visits.

Evaluation of anxiety (STAI-Y scale)

The STAI questionnaire is a tool used to detect and mea-sure anxiety. It is employed for psycho-diagnostic purposes as well as for verifying the effectiveness and benefits of psychotherapy.16STAI is available in 2 versions: X15and Y (STAI-Y, evolution of the X version).16 In this study, the level of anxiety of the subjects was measured using the updated and validated Italian version of STAI-Y.17 The STAI-Y questionnaire is a self-report instrument which consists of 40 statements (items) to which the sub-jects must respond with an intensity score (on a 4-point Likert scale, from “almost never” to “almost always” or from “not at all” to “a lot”). The items are grouped into 2 self-report subscales (STAI-trait and STAI-state), relating respectively to how people generally feel (trait anxiety), and to how they feel in certain situations (state anxiety). State anxiety denotes an individual’s transient emotional response to a particular situation, manifested in particular as a feeling of insecurity or powerlessness in the face of a perceived damage or danger, which can trigger apprehen-sion or a desire to flee. Trait anxiety, on the other hand, re-flects the stable component of personality that makes some individuals more susceptible than others to experience high levels of anxiety in stressful situations perceived as dangerous or threatening.15,16

Scores on each subscale can range from 20 to 80, with higher scores corresponding to greater anxiety. According to Spielberger et al.,15 anxiety can be classified as low (score 20e30), moderate (score 31e49), and high or severe (score 50e80). The median is commonly used as the dividing line between low- and high-anxiety patients. Both the state and trait components of the STAI-Y ques-tionnaire were administered to each subject at both visits.

Sleep quality: Sleep quality was assessed through self-reported average number of hours slept per night (“sleep hours”) and the average number of awakenings per night. At the first visit (V1), subjects were asked to describe their usual sleep pattern in terms of both average number of hours slept and average number of night-time awakenings. These estimates reported by patients at the first visit were considered the baseline value. At the follow-up visit (V2), they were again asked to answer the same questions about the average sleep patterns they had experienced dur-ing the period between the 2 visits.

The number of night-time awakenings was coded in a range from 0.5 (few night-time awakenings, not occurring every night) to 9 (many awakenings or no sleep at all). In addition, to reduce data dispersion, we performed an analysis by arbitrarily grouping subjects into 3 groups: (1) subjects with few or no awakenings (from 0 to 2 awak-enings per night, not necessarily occurring every night);

(2) subjects with some awakening (up to 4 awakenings per night); (3) subjects with many awakenings (over 4 awakenings per night).

Statistical analysis: The demographic characteristics of the study population were analyzed and reported as fre-quencies and percentages (gender, age class, anxiety level) and by mean, standard deviation (SD), range (age). Given the retrospective nature of this study, the sample size was not estimated a priori. The study population was drawn from patients who had been referred to the Department of Dietary Nutrition and Complementary Medicine, “Azienda Ospedaliera Gaetano Rummo”, Benevento, Italy, in the period between January 2010 and December 2010 and who were prescribed homeopathic treatment for anxi-ety. A sample size of 100 subjects was estimated to be suf-ficient to provide a statistically significant difference of 3 points (SD of 6 points) from the STAI-Y score before and after treatment, with 90% power (a = 0.05).

The STAI-Y scores at Visit 1 and Visit 2 were analyzed considering the mean, SD, and range, and compared using the t-test for repeated measures. The median of the distri-bution of STAI-Y values at Visit 1 was taken as the cut-off point to distinguish between low-anxiety and high-anxiety subjects: this corresponded to 44 points for the STAI-state scale and 48 points for the STAI-trait scale. Stratified analysis by gender, age group and anxiety level was also performed.

The characteristics related to sleep were examined by comparing the “sleeping hours” and the number of night-time awakenings reported during the 2 visits using the Wilcoxon signed-rank test, since these variables were not normally distributed. The number of night-time awak-enings was categorized as a categorical ordinal variable and analyzed by the chi-squared test.

Results

Demographic and baseline characteristics

A total of 71 subjects attended both visits, of whom 70% were female (n = 50). On average, subjects had a second visit 2 months after treatment was started. The average age of the subjects was 52 years (range: 29e81 years), and more than half were over 50 years old. The subjects were classified as having low or high-anxiety levels based on the median value. However, in the STAI-trait subscale, a number of subjects had values close to the median, causing a little unbalancing (Table 1).

Evaluation of anxiety (STAI-Y scale)

The level of anxiety was measured at baseline (V1) and follow-up (V2) using the STAI-state and STAI-trait scores (Table 2).

At the baseline visit (V1), the mean STAI-state score was 43.61 and the mean STAI-trait score was 47.89 (Table 2): both mean scores were within the moderate anxiety range (31e49).15,16Both scores were reduced at the follow-up visit (V2, occurring after an average time of 2 months), and the difference was statistically significant for both in-dicators (p < 0.001). The mean STAI-state score decreased

by more than 6 points, and the mean STAI-trait scale by more than 3 points (Table 2). A few subjects ran counter to this trend: the STAI-state score increased slightly (by 1e4 points) between V1 and V2 for 6 subjects, STAI-trait increased slightly (by 1e4 points) between V1 and V2 for 7 subjects, and one of the 71 participants showed an increase in both subscales at the second visit. At base-line (V1), women had higher scores compared to men for both subscales, but the average decrease at V2 was also

higher among women. The difference between scores at V2 and V1 was always found to be statistically significant, for both sexes and both scales (Table 2).

Considering the wide age range of the subjects (29e90), we also performed an age-stratified analysis of the STAI-state and STAI-trait scores. The differences between the scores recorded at V2 compared to those recorded at V1 were found to be statistically significant within each age group, even when the number of subjects was relatively small, such as in the younger age group (p < 0.01) (Table 2). The higher anxiety group showed a notable decrease in Y scores for both subscales (10 points in STAI-state and 4 points in STAI-trait) while the lower-anxiety group had a reduction of approximately 2 points for both subscales (Table 2). Between the first and the second visit, 29 out of 38 subjects (76.32%) switched from high state anxiety level to low state anxiety level and 19 out of 43 sub-jects (44.19%) switched from high trait anxiety level to low trait anxiety level.

Sleep duration

Subjects reported having had trouble sleeping on average for the past 2.5 years, with a minimum of 15 days and a maximum of 10 years. One subject reported having always had trouble sleeping.

Between the first and second visit, an average increase of 28.2 min of sleep per night was reported by the subjects, the difference was statistically significant (Figure 1A;

Table 1 Demographic and baseline characteristics Characteristics Values Gender, N (%) Women 50 (70.42) Men 21 (29.58) Age, years Mean age (SD) 52.69 (12.52) Range 29e81 Age class, N (%) <40 years 10 (14.08) 40e49 years 22 (30.99) 50+ 39 (54.93) State anxiety, N (%) Low (STAI-state <44) 33 (46.48) High (STAI-state$44) 38 (53.52) Trait anxiety, N (%) Low (STAI-trait <48) 28 (39.44) High (STAI-trait$48) 43 (60.56) Number (N) and percentage (%) of subjects per category and mean, SD and range of age are reported.

Table 2 STAI-state and STAI-trait scores at first (V1) and second (V2) visit

STAI-Y Visit N Mean SD Min Max Mean difference p-value Cohort State V1 71 43.61 9.18 22 68 6.37 <0.0001

V2 37.24 5.54 22 53

Trait V1 47.89 7.00 34 62 3.31 <0.0001 V2 44.58 6.25 33 57

Gender Women State V1 50 44.02 9.07 26 62 6.98 <0.0001 V2 37.04 5.36 22 48 Trait V1 48.58 6.56 36 62 4.90 0.0012 V2 45.12 5.95 35 55 Men State V1 21 42.62 9.59 22 68 3.46 <0.0001 V2 37.71 6.04 25 53 Trait V1 46.24 7.87 34 60 2.95 <0.0001 V2 43.29 6.89 33 57

Age class <40 State V1 10 45.70 9.93 31 68 8.70 0.0014 V2 37.00 6.29 30 53 Trait V1 50.30 5.98 40 60 5.50 0.0020 V2 44.80 5.12 37 57 40e49 State V1 22 43.14 10.46 22 61 6.45 0.0001 V2 36.68 6.08 22 46 Trait V1 48.23 6.43 34 58 3.82 <0.0001 V2 44.41 6.01 35 55 50+ State V1 39 43.33 8.38 30 62 5.71 <0.0001 V2 37.62 5.13 29 48 Trait V1 47.08 7.52 34 62 2.46 <0.0001 V2 44.62 6.77 33 55

Anxiety level Low State V1 33 35.70 5.26 22 43 2.18 0.0020 V2 33.52 4.00 22 42 Trait V1 28 40.57 4.13 34 47 2.00 0.0002 V2 38.57 3.44 33 47 High State V1 38 50.47 5.60 44 68 10.00 <0.0001 V2 40.47 4.59 29 53 Trait V1 43 52.65 3.42 48 62 4.16 <0.0001 V2 48.49 4.22 40 57

Mean, SD, range (MineMax), mean difference and t-test p-values are reported for the whole cohort and by gender, age class and anxiety level subgroups.

p < 0.0001). The duration of sleep between the first and the second visit remained unchanged in 34% of patients, improved in 62% of them, and decreased only in 4% of the study population.

Number of awakenings

A marked decrease in the number of awakenings per night was observed between the first and second visit (Figure 1B; 42%, p < 0.0001). The number of awaken-ings decreased in 66% of patients, remained unchanged in 20% and increased only in 14% of subjects. In addition, further examination of grouped data revealed a clear decrease in the number of subjects with >4 awakenings and an increase in subjects with 0e2 awakenings; (Figure 1C).

Safety and concomitant medication

No adverse events were reported among the 71 patients who were treated with Datif-PCÒ.

Three subjects reported having taken another medication before the first visit. One subject had taken a herbal medi-cine (“Erba della nottee Night Herb”), another one Clo-nazepam, and the third had taken Lamotrigine. One

subject had taken a tablet of Ropinirole Hydrochloride be-tween the 2 visits.

Discussion

The findings of this retrospective observational study suggest that treatment of anxiety with Datif-PCÒcan bring about notable improvements in a relatively short period of time, in terms of better STAI scores, increased sleep dura-tion and fewer nocturnal awakenings in patients previously diagnosed with mild to moderate anxiety.

The aim of our study was to investigate how the 2 com-ponents of anxiety (trait and state) changed over time in a cohort of 71 subjects treated with the homeopathic-complex medicine Datif-PCÒ. We also assessed changes in the sleep quality of the study population through analysis of the number of “sleeping hours” and the number of nocturnal awakenings.

The natural course of generalized anxiety disorders is characterized as a chronic condition with few remissions, a waxing and waning progression, with substantial inci-dence of comorbidity including, but not limited to, depres-sion, alcohol abuse, and other anxiety disorders. The probability of remission of generalized anxiety disorders is only 38% at 5 years, and the probability of relapse after remission is 27% by 3 years. There is a growing awareness of the disability and impaired quality of life associated with anxiety disorders.18

To date, few studies have specifically examined the treat-ment of anxiety in the general population using homeopath-ic medhomeopath-icines. Although individuals suffering from anxiety frequently turn to homeopathic medicines,11e14 the studies conducted thus far cannot confirm the efficacy of such treatments. The findings for the primary endpoint of this study, concerning the change in STAI-Y scores over time, suggest that Datif-PCÒmay be effective in controlling anxiety: the scores decreased overall for both the STAI-state and STAI-trait subscales. A systematic review of 8 ran-domized controlled studies concluded that no firm conclu-sion can be drawn concerning the efficacy or effectiveness of homeopathy for anxiety, but that if homeopathy were shown to be effective, it would likely have benefits in terms of fewer adverse events and better acceptability for pa-tients.11 Recently, another systematic review performed on 25 homeopathy studies found evidence of efficacy for functional somatic syndromes but not for anxiety or stress, while the safety data e despite being superficially collectede suggested good tolerability of homeopathy.19 A more recent review found that certain lifestyle modifica-tions and certain complementary medicines and therapies may have a beneficial role in the treatment of anxiety disor-ders, but found no support for the use of homeopathy to treat anxiety disorders.20 A double-blind, placebo-controlled, randomized clinical trial with 3 treatment groups found that homeopathic Argentum nitricum 12 (both tradition-ally and radionictradition-ally prepared) did not reduce test anxiety in a general population of 62 university students.21A pro-spective, research-targeted pilot study, in which 14 homeo-pathic physicians collected clinical and outcome data over a

Figure 1 Sleep duration and nocturnal awakenings reported by subjects (n = 71) at first (V1) and second visit (V2). (A) Number of hours (h) slept per night. (B) Number of nocturnal awakenings. (C) Grouped number of nocturnal awakenings. Data presented as mean and SD for (A) and (B). Data presented as percentage of pa-tients for (C). Wilcoxon signed-rank sum test p-values are re-ported.

6-month period in their practice setting, found strongly pos-itive outcomes were achieved most notably in the frequently-treated conditions of anxiety and depression.22 A double-blind randomized trial was carried out to compare the efficacy of the homeopathic medicines Homeogene 46 and Sedatif PC (identical to Datif-PCÒ, used during this observational study) with a placebo as substitute for benzo-diazepines: no statistically significant differences were found, but the study did suffer from a lack of statistical po-wer, due to recruiting difficulties.23

From this retrospective study, Datif-PCÒemerges as hav-ing been most effective in controllhav-ing anxiety in subjects who presented with high STAI-Y scores at the first visit. Those individuals in fact showed marked reductions in anx-iety scores (10 points in state and 4 points in STAI-trait) after just 2 months of treatment on average. The decrease in the average score on the STAI-state subscale was higher than that on the STAI-trait scale. This is to be ex-pected due to the more transient character of state anxiety, which is an emotional response elicited by a particular situ-ation or stimulus, that causes an individual to feel insecure or powerlessness in the face of a perceived threat or harm, thus eliciting apprehension or a desire to escape. In contrast, trait anxiety is a more stable feature of an individual’s per-sonality, that determines his or her propensity to respond with high levels of anxiety to stressful situations perceived as dangerous and threatening.16 Thus, since state anxiety is a more transitory condition, it is more susceptible to modi-fication than trait anxiety, which instead reflects a stable aspect of a person’s personality.16

The results observed for the entire cohort are mirrored in the subgroup analysis. In fact, STAI-state scores showed a greater decrease compared to the STAI-trait scores in each subgroup, both by sex and age class. In particular, subgroup analysis showed that, at the first visit, women had higher average scores than men for both subscales, and that after treatment they reported a more significant decrease in those scores. Clustering by age class showed that young people on average had higher values at the first visit on both scales, but that they also experienced a more substantial decrease in anxiety at the second visit, which was statistically signif-icant despite the small sample size (n = 10).

The results of this study are noteworthy, first of all because a statistically significant improvement was ob-tained, even with a relatively short treatment period and a restricted number of subjects. A second important point is that, with respect to the Italian population17, the mean scores recorded at the second visit of the study can be placed at least 5 percentiles below those recorded at the first visit, for the relevant subscale and subgroup.

Sleep quality was examined by analyzing the number of hours slept per night and the number of night-time awaken-ings. Both these variables showed a statistically significant improvement, although the average increase in the number of hours slept was about half an hour, the majority of sub-jects (84%) reported an increase in the period slept per night and/or a decrease in the number of nocturnal awaken-ings: 44% of subjects experienced an improvement for both indicators. Bearing in mind that the subjects included in

this study reported having experienced sleep disorders for the past 2.5 years on average (maximum of 10 years), these results are encouraging.

A meta-analysis of the use of benzodiazepines for the treatment of insomnia24found that benzodiazepines have a greater efficacy in terms of increased sleep duration (about 1 h increase per night), but a serious array of adverse effects. Considering the lack of any adverse events reported by the patients in this study, the risk-benefit ratio favors the use of the homeopathic medicine rather than benzodiazepines for the treatment of sleep disorders. An updated systematic re-view on the use of homeopathy for treating insomnia found significant improvements in the groups treated with home-opathy relative to baseline, though no statistically signifi-cant difference compared to the placebo-treated group. This lack of statistical significance was prevalently due to the lack of statistical power of the study. In addition, the studies were of low methodological quality and poorly re-ported.25,26 A randomized clinical trial of higher quality, i.e., with appropriate statistical power and clear reporting of methodological details, found significant improvement in the homeopathy group with respect to the placebo group, in addition to improvements from baseline.27

Limitations

This was an open-label, retrospective observational study whose design differs from a clinical trial protocol. Its aim was to describe the clinical practice ordinarily fol-lowed by the clinic where the study was performed, over the course of 1 year.

Homeopathic treatment generally implies a preliminary consultation aimed to assess the general psychophysical condition of the patient. An in-depth consultation, including psychological consultation and advices on life-style, could contribute to an improvement of anxiety and sleep disorder. In this study, the psychological support was limited to the assessment of the disorders through the administration of questionnaires and no advice was given on personal lifestyle. For this reason, the impact of factors other than the homeopathic medicine should have been reduced at the minimum possible.

No direct assessment of compliance was possible, but all subjects at V2 stated that they had followed the physician’s prescription. In addition, the lack of any comparison with another treatment, whether homeopathic or conventional, is weakness of this study design, as is also the lack of com-parison with a placebo. To partially address this issue we compared our results with those of the studies mentioned above, keeping in mind 2 points:

1. the limitations of both the present study and the other studies

2. the comparability between the present study and the other studies not properly assessed

The percentage of subjects reporting improved sleep pattern in the placebo groups in the studies reported above25e27 ranged between 0 and 50%, well below the percentage observed in our study (84%). Looking at the

mean duration, Naude et al.27found an improvement of one hour per week in the placebo group, while in our study the mean improvement was about half an hour per day (i.e., about 3.5 h per week). Such a comparison was not feasible for the anxiety measures.

The lack of comparison could have allowed an exagger-ated impact of the expectation bias on the result. However, in this study the expectation bias should not be relevant, as results have been collected through both validated ques-tionnaires administered repeatedly and self-reported quan-titative measure (such as the number of awakenings and of hours slept per night).

As a final limitation, the retrospective design meant it was not possible to establish a priori the duration of treat-ment and schedule of visits. Our approach was to include data from patients who had been treated for at least one month: the treatment periods ranged from 31 to 71 days.

Conclusion

The findings of this study suggest that the homeopathic-complex medicine Datif-PCÒ may provide favorable ef-fects, with a good safety profile, in the absence of concom-itant therapies. The studied medicine does not share the unfavorable side effects associated with benzodiazepines, which are frequently prescribed for the treatment of insomnia and anxiety disorders. Though subject to the described limitations, this retrospective observational study suggests that treatment of anxiety and sleep disorders with a homeopathic-complex medicine may help bring about im-provements in a relatively short period of time. Further-more, the observed effects were independent of gender, age or baseline anxiety level. According to the previously cited reviews,11,19,20 nearly all previously published studies of homeopathy in this clinical domain suffer from major methodological flaws (e.g. small samples, lack of control group), so that their quality is insufficient to support the use of homeopathy to treat anxiety disorders, despite some results suggestive of its efficacy.22,23 The present study is, likewise, at least suggestive of a positive effect of Datif-PCÒ: the improvement in STAI-Y scores was both clinically and statistically significant. However the lack of a control group means we cannot draw firm con-clusions. For this reason, further studies are needed to confirm these findings. These studies should be randomized controlled clinical trials, of adequate sample size, based on validated self-reported outcome for both anxiety and sleep disorders, as partially done in this study. Objective mea-surements of sleep quality and quantity should also be considered (for instance, wearable armband sensors).

Conflict of interest statement

The authors have no conflict of interest to declare.

Acknowledgments

The authors wish to thank Dr Colin Gerard Egan and Dr Selene Mogavero (Primula Multimedia S.r.L., Pisa,

Italy) for revising the English of the manuscript and Dr Ma-riacarla Di Gioia, Angela Rita Rotondi and Lara Rossi for their advice and collaboration.

References

1 American Psychiatric Association. Diagnostic and statistical manual of mental disorders. text revision. 4th ed. Washington, DC: American Psychiatric Pub, 2000.

2 National Institute of Mental Health. Anxiety disorders. NIH Publi-cation No. 09 3879; 2009.

3Demyttenaere K, Bruffaerts R, Posada-Villa J, et al., for the WHO World Mental Health Survey Consortium. Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization World Mental Health Surveys. JAMA 2004; 291: 2581e2590.

4Taylor DJ, Lichstein KL, Durrence HH, Reidel BW, Bush AJ. Epidemiology of insomnia, depression, and anxiety. Sleep 2005; 28: 1457e1464.

5 Van Mill JG, Hoogendijk WJ, Vogelzangs N, van Dyck R, Penninx BW. Insomnia and sleep duration in a large cohort of pa-tients with major depressive disorder and anxiety disorders. J Clin Psychiatry 2010; 71: 239e246.

6 Hartz AJ, Daly JM, Kohatsu ND, Stromquist AM, Jogerst GJ, Kukoyi OA. Risk factors for insomnia in a rural population. Ann Ep-idemiol 2007; 17: 940e947.

7 Lader M. Benzodiazepines revisitedewill we ever learn? Addiction 2011; 106: 2086e2109.

8Koen N, Stein DJ. Pharmacotherapy of anxiety disorders: a critical review. Dialogues Clin Neurosci 2011; 13: 423e437.

9 Rudolph U, Knoflach F. Beyond classical benzodiazepines: novel therapeutic potential of GABAA receptor subtypes. Nat Rev Drug Discov 2011; 10: 685e697.

10Kripke DF, Langer RD, Kline LE. Hypnotics’ association with mor-tality or cancer: a matched cohort study. BMJ Open 2012; 2: e000850.

11Pilkington K, Kirkwood G, Rampes H, Fisher P, Richardson J. Ho-meopathy for anxiety and anxiety disorders: a systematic review of the research. Homeopathy 2006; 95: 151e162.

12Trichard M, Lamure E, Chaufferin G. Study of the practice of ho-meopathic general practitioners in France. Homeopathy 2003; 92: 135e139.

13Pedersen EJ, Norheim AJ, Fønnebe V. Attitudes of Norwegian phy-sicians to homeopathy. A questionnaire among 2019 phyphy-sicians on their cooperation with homeopathy specialists. Tidsskr Nor Laege-foren 1996; 116: 2186e2189.

14Grolleau A, Begaud B, Verdoux H. Characteristics associated with use of homeopathic drugs for psychiatric symptoms in the general population. Eur Psychiatry 2013; 28: 110e116.

15Spielberger CD, Gorsuch RL, Lushene RE. Manual for the State-Trait Anxiety Inventory (self-evaluation questionnaire). Palo Alto, CA: Consulting Psychologists Press, 1970.

16Spielberger CD. Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press, 1983.

17Spielberger CD. STAI State-Trait Anxiety Inventory form Y. Firenze: Edizione Organizzazioni speciali, 1996.

18Huh J, Goebert D, Takeshita J, Lu BY, Kang M. Treatment of gener-alized anxiety disorder: a comprehensive review of the literature for psychopharmacologic alternatives to newer antidepressants and benzodiazepines. Prim Care Companion CNS Disord 2011; 13(2). PCC.08r00709.

19Davidson JR, Crawford C, Ives JA, Jonas WB. Homeopathic treat-ments in psychiatry: a systematic review of randomized placebo-controlled studies. J Clin Psychiatry 2011; 72: 795e805. 20Sarris J, Moylan S, Camfield DA, et al. Complementary medicine,

exercise, meditation, diet, and lifestyle modification for anxiety 260

disorders: a review of current evidence. Evid Based Complement Al-ternat Med 2012; 2012: 809653.

21Baker DG, Myers SP, Howden I, Brooks L. The effects of homeo-pathic Argentum nitricum on test anxiety. Complement Ther Med 2003; 11: 65e71.

22Mathie RT, Robinson TW. Outcomes from homeopathic prescribing in medical practice: a prospective, research-targeted, pilot study. Homeopathy 2006; 95: 199e205.

23Cialdella P, Boissel JP, Belon P, Groupe de recherche ASTRHO. Homeopathic specialties as substitutes for benzodiazepines: double-blind vs. placebo study. Therapie 2001; 56: 397e402.

24Holbrook AM, Crowther R, Lotter A, Cheng C, King D. Meta-anal-ysis of benzodiazepine use in the treatment of insomnia. CMAJ 2000; 162: 225e233.

25Cooper KL, Relton C. Homeopathy for insomnia: a systematic re-view of research evidence. Sleep Med Rev 2010; 14: 329e337. 26Cooper KL, Relton C. Homeopathy for insomnia: summary of

addi-tional RCT published since systematic review. Sleep Med Rev 2010; 14: 411.

27Naude DF, Couchman IMS, Maharaj A. Chronic primary insomnia: efficacy of homeopathic simillimum. Homeopathy 2010; 99: 63e68.