The long-term effectiveness of CBT is still the objective of future research. The effectiveness of psychological therapy is compared with the effectiveness of pharmacological treatment, and the conclusions have determined not that CBT does not have lasting effects, but rather that more studies based on a robust methodology and carried out with reliable data are needed¹⁰⁷.

Sufficient evidence was not found in trials researching symptomatic treatment of panic attacks, especially in terms of acute treatment of the crisis^70,80-82.

The Canadian guide emphasizes the role of cognitive-behavioral therapy as an effective psychotherapeutic strategy for handling anxiety crises⁶¹. In our context, and within the scope of Primary Care, behavioral and support measures that contain psycho-education (calming the patient, written advice on actions), training in handling the symptoms (instruction in relaxation techniques and learning breathing exercises to handle hyperventilation), and exposure techniques could also be included. It is also important to inform the family regarding this type of actions in response to the appearance of a new crisis.

Independently from the focus from which one works, and whether the treatment is done individually or as a group, psychological interventions in PC must be done by trained professionals and must have common characteristics of applicability⁷⁷, which are the characteristics which differentiate them from the exercise of the support relationship that is created in Primary Care facilities:

1. Structured: simple to learn and easy to apply.
2. Brief and with specified times in the initial manualization.
3. With specific objectives established after evaluation.
4. Effectiveness described for the diagnosis to be treated.

These techniques help to understand the symptoms and underlying conflicts (from the internal world or personal relations) through an empathic focus in which the therapist observes, comprehends, and receives the person’s anxieties, and returns them so that they can be recognized and accepted by the patient, encouraging thinking and taking responsibility for themselves⁷⁸.

This therapy understands the family as a set of elements that interact with each other and that cause people to persist in activities that keep the problems alive. The goal of the professional must be to stimulate to initiate a process of change. A proposal is formulated for actions that foster the desired changes to resolve the problem. The patient and the patient’s family complete the process, without accompanying them to the end of the solution and without the entire family visiting the doctor.

Counseling Techniques (assisted counseling)^59,117

Use of relationships that develop self-awareness, emotional acceptance, growth, and personal resources. This may be directed to resolve specific problems, to make decisions, confronting a crisis, working through the feelings within conflicts, or improving relations with others. The problem is explored with the patient from effective comprehension to allow the patient to establish goals and objectives.

Most of these techniques, both cognitive-behavioral and psychodynamic, can be applied individually or in groups, provided that the intervention fulfills the requirements described earlier and is properly structured for application in a group format.

The most commonly used group therapies used in Primary Care for anxiety disorders are those called “Skill development”. They are used to apply the learning of breathing and relaxation techniques, facing and handling stress, resolving problems, and training and handling of anxiety, among others.

One of the first systematic reviews of RCTSs that use anti-depressants (imipramine, paroxetine, and trazodone) to treat anxiety disorders showed comparable effectiveness between benzodiazepines and anti-depressants for acute treatment of GAD¹²². The data gathered over the course of the last two decades has continued to highlight the effectiveness of anti-depressants in the treatment of GAD.

When the effectiveness of the anti-depressants imipramine, venlafaxine, and paroxetine and their degree of acceptance, measuring the results in terms of “absence of response” (the response defined as the absence of symptoms sufficient to fulfill the diagnostic criteria, with ratings of 1 or 2 – very significant improvement – in the Clinical Global Impressions – CGI), “abandonment rate”, and “specific side effects”, were studied against a placebo, it was observed that there is a greater probability of response to the treatment in the short term in the case of anti-depressants versus the placebo, with a global NNT for anti-depressants of 5.5 (CI of 95%:4.1;8.4), that there were no significant differences in terms of abandonment between the two, and that the side effects manifested themselves most frequently in the groups treated with drugs versus those treated with the placebo. The fact that the abandonment rates did not show significant differences between the groups treated with anti-depressants and those treated with the placebo suggests that patients with GAD can tolerate the use of these drugs well. When tricyclic anti-depressants are compared with new anti-depressants, the results in terms of effectiveness and tolerability are similar for paroxetine and imipramine. Although venlafaxine and paroxetine are associated with a better level of acceptance, no differences were found with tricyclic imipramine in terms of abandonment, which is possibly a more solid indicator of the level of acceptance^123,124.

Another systematic review and later trials expanded the evaluation of the aforementioned anti-depressants with sertraline, escitalopram, and opipramol*, comparing them with a placebo. It was observed that these drugs increase response rates and improve symptoms of GAD^87,125. Research with other anti-depressants for treating GAD highlight the role of other drugs, such as duloxetine, for its good tolerance levels and effectiveness in comparison with a placebo to reduce functional alterations of patients and improve quality of life and well-being. More studies are needed on its effectives in comparison with other anti-depressants^126-128.

The adverse effects found for the aforementioned families of anti-depressants are associated with sedation, dizziness, nausea, dry mouth, constipation, falls, and sexual dysfunction, among others, although many of the trials does not reflect their statistical significance. However, there is evidence that demonstrates that most of them (with the exception of dizziness and sexual dysfunction) decrease after 6 months in patients who continued with the medication. On the other hand, the sudden interruption of treatment with SSRIs is associated with adverse effects such as dizziness, headache, nausea, vomiting, diarrhea, movement disorders, insomnia, irritability, visual alterations, lethargy, anorexia, and states of despondency. The possible increase in the risk of self-inflicted injuries, suicide, and hyponatremia must be taken into consideration in regard to the use of SSRIs.^87,129.

The US FDA (Food and Drug Administration) has warned of complications when taken during pregnancy: congenital malformations, especially cardiac malformations, if paroxetine is used in the first trimester of pregnancy, increasing its pregnancy risk category from C to D (the FDA’s classification for medications based on their teratogenic potential)^87,130. Also, when SSRIs are taken in the final phases of pregnancy, there is some evidence that indicates that these drugs interfere with the respiratory and parasympathetic systems of neonatal infants, along with increased risk of respiratory and central nervous system symptoms. Hypoglycemia and neonatal adaptation problems may also be encountered. However, all of these results are not always due to the toxicity or removal of SSRIs. To reduce the potential risk of adverse neonatal effects, the lowest effective dose of SSRIs should be used with the shortest possible treatment duration, as monotherapy whenever feasible. During pregnancy, the choice of the treatment must consider whether the potential advantages for the mother of the prescribed SSRIs outweigh the possible risks to the fetus^131-136.

The guides from the NICE, Canadian Psychiatric Association, and MOH^70,80-82, consider the use of anti-depressants to be one of the treatments of choice for GAD. The Canadian guide specified paroxetine, escitalopram, sertraline, and slow-release venlafaxine as the best options for pharmacological treatment because of their significant improvements in quality of life and the symptoms related to functional disability^70,137. Paroxetine, escitalopram, and venlafaxine have demonstrated long-term effectiveness, with response rates that continue to increase beyond 6 months of treatment. For patients who interrupt treatment, there is a risk of relapse from 20% to 40% between 6 and 12 months after treatment is interrupted. This therefore suggests that long-term treatment will often be necessary^70,138-140.

Warning regarding venlafaxine: the NICE guide indicates that this drug has a higher probability of interruption of treatment due to side effects and higher cost than SSRIs, compared with the same level of effectiveness. It also includes a warning based on evidence provided by the MHRA (Medicines and Healthcare products Regulatory Agency). This agency has warned against the cardio-toxic and hypertensive effects of this drug, especially associated with higher-than-therapeutic dosages (a dosage not exceeding 75 mg/day is recommended). NICE recommends that when venlafaxine is prescribed to patients with hypertension, that the hypertension be controlled and that it not be prescribed to patients with a high risk of cardiac arrhythmia or recent heart attack^41,141.

When the response to the optimal dosage of one of the SSRIs is inadequate or if SSRIs are not well tolerated, the patient should switch to another SSRI. If there is no improvement after 8-12 weeks, consider using another drug with a different mechanism of action (NSRI, TAD)^70,80.

BDZs are part of a family of drugs that have demonstrated their effectiveness in the treatment of generalized anxiety^70,80-82.

Alprazolam, bromazepam, lorazepam, and diazepam have been proven to be effective in the treatment of GAD, and the Canadian guide includes them as the second-tier pharmacological treatment. There is insufficient evidence evaluating the effectiveness of clonazepam, with long average life and low potential for rebound anxiety, but it is probable that benefits similar to those of other BDZs will be obtained⁷⁰.

BDZs provide fast initial relief of anxiety symptoms, but the evidence suggests that their effects do not differ significantly from those obtained with a placebo after 4 to 6 weeks of treatment. Also, BDZs primarily reduce somatic symptoms more than psychic symptoms (apprehension), which are the symptoms that define GAD⁷⁰.

Side effects have been observed with the use of benzodiazepines in terms of increased risk of dependency, tolerance, sedation, traffic accidents, and effects of suspension of treatment (rebound anxiety)⁸⁷. Existing evidence is insufficient to determine whether during pregnancy, the potential benefits of BDZs for the mother outweigh the possible risks to the fetus^142-144. To avoid the potential risk of congenital defects, the lowest effective dosage of BDZs should be used, with the shortest possible treatment duration, and as monotherapy. If higher concentrations are required, the daily dosage should be divided into two or three doses, always avoiding use during the first trimester^143-145. In advanced stages of pregnancy or during nursing, BDZs can cause adverse effects in neonatal infants (neonatal hypotonia, withdrawal syndrome, sedation, and hypothermia)^{82,87,146-149}.

Due to its effectiveness and the adverse effects described, short-term use is recommended, not extending beyond 2 to 4 weeks, especially when rapid control of the symptoms is required and while waiting for response to the benefits of treatment with anti-depressants or CBT. Long-term use must be closely supervised^70,80-82.

Azapirones are a family of anxiolytic drugs that act on the 5-HT1A receptor. The effectiveness and acceptability of azapirones (buspirone) compared with a placebo or other treatments has been evaluated, and they appear to be useful and better than the placebo over the short term (four to nine weeks) in treating GAD, especially if patients have not taken benzodiazepines before. It is not possible to conclude whether azapirones are better than benzodiazepines, anti-depressants, psychotherapy, hydroxicine, or the extract from the kava kava plant. The side effects do not appear to be serious and they involve physical symptoms (nausea, dizziness, and drowsiness) above all^70,145,150.

Since GAD is generally chronic in nature, more studies must be done to establish conclusions regarding its long-term effectiveness.

Although azapirones have been approved for treatment of GAD in Spain, their use is very limited.

Pregabaline in an anti-convulsive that when compared with a placebo, has been proven effective against the psychic and somatic symptoms of GAD, and is also tolerated by the majority of patients. The adverse effects associated with it include drowsiness, vertigo, and headache. Clinical experience with this drug is limited^{70,87,151-154}.

Hydroxicine is a medication derived from piperidine, used generally as an anti-histamine. It also acts as a sedative and tranquilizer, which makes it useful for treating anxiety.

When hydroxicine was compared with a placebo, it was found to improve anxiety symptoms. When compared with other drugs, such as bromazepam and buspirone, no significant differences in effectiveness were observed. The side effects found are, above all, headache and drowsiness. Clinical experience with this drug is also limited^70,87.

Open trials with small sample sizes suggest that the atypical anti-psychotics olanzapine, risperidone, and zyprasidone can have some benefit as adjuvant drugs in the treatment of refractory GAD, although controlled trials with placebos, randomized double-blind studies, and greater strength are required to verify its effectiveness and safety^70,155,156.

Other anti-depressants such as mirtazapine, zitalopram, trazodone, and slow-release bupropion, anti-convulsives such as tiagabine, the drug used in treating Lateral Amyotrophic Sclerosis (ryluzol), and new anxiolytics like deramcyclane may show some effectiveness in treating GAD^70,157-159.

RCTS with placebos and larger sample sizes are needed to confirm these results.

Beta-blockers such as propranolol have not been found to be more effective than placebos in treating GAD^70,80.

One of the first meta-analyses with RCTS that used anti-depressants to treat PD demonstrated the effectiveness of SSRIs for this disorder¹⁶³.

Along this same line, later reviews that analyzed the effectiveness of anti-depressants found that paroxetine, fluoxetine, fluvoxamine, citalopram, sertraline, chlorimipramine, and imipramine improved the symptoms of PD in comparison with a placebo. The adverse affects associated with these drugs are headache, trembling, dry mouth, drowsiness, nausea, and dizziness, among others. The percentage of abandonment due to adverse effects was 11% and was similar among the SSRIs and TADs^98,164.

The US FDA (Food and Drug Administration) has issued several warnings in regard to the use of SSRIs and increased risk of self-inflicted injury, suicide, and hyponatremia¹²⁹. It has also warned of complications when taken during pregnancy: persistent neonatal pulmonary hypertension and congenital malformations, especially cardiac malformations, if paroxetine is used in the first trimester of pregnancy, increasing its pregnancy risk category from C to D (the FDA’s classification for medications based on their teratogenic potential)¹³⁰. Also, when SSRIs are taken in the final phases of pregnancy, there is some evidence that indicates that these drugs interfere with the respiratory and parasympathetic systems of neonatal infants, along with increased risk of respiratory and central nervous system symptoms. Effects of hypoglycemia and adaptation problems may also be observed. However, all of these results are not always due to the toxicity or removal of SSRIs. To prevent potential risk of adverse neonatal effects, the lowest effective dose of SSRIs should be used with the shortest possible treatment duration, with the possibility of use as monotherapy^131-136 .

The guides from the NICE, Canadian Psychiatric Association, and MOH consider the use of SSRI anti-depressants to be the pharmacological treatment of choice for PD^70,80-82.

The Canadian guide specifies citalopram, escitalopram, fluoxetine, fluvoxamine, paroxetine, sertraline and slow-release venlafaxine as the best choices for pharmacological treatment due to their significant improvement in the gravity of the panic. The aforementioned SSRIs also show a significant improvement in anticipatory anxiety and in agoraphobic avoidance, as well as in the symptoms related to functional disability and quality of life. Paroxetine, citalopram, fluoxetine, sertraline, and venlafaxine have demonstrated prolonged benefits and continued improvements over the course of 6 to 12 months of treatment^70,165,166.

In terms of the use of venlafaxine, the MHRA (Medicines and Healthcare products Regulatory Agency) has warned of the cardio-toxic and hypertensive effects of venlafaxine, especially associated higher-than-therapeutic doses¹⁴¹.

There is limited evidence in regard to long-term persistence (longer than 6 months after treatment has been completed) of the benefits obtained with short-term treatments when the continuation of the treatment is not maintained¹⁶⁷ maintained167 and it is difficult to establish the ideal duration of pharmacological treatment for PD. Although imipramine and venlafaxine have been proven to prevent long-term relapse in comparison with the placebo, the interruption of treatment with anti-depressants poses a risk of relapse, so therapy in many patients should be applied long-term (at least 12 months)^70,168.

The treatments administered during the studies and their follow-up period must be better supervised in order to obtain more data on the long-term results.

Research with other anti-depressants in the treatment of PD have demonstrated the effectiveness of mirtazapine and milnacipran (anti-depressant serotonin-5HT and noradrenalin-NA reuptake inhibitor) in open trials^70,169.

Other studies with MAOIs and MARIs show that some, such as fenelzine, appear to be effective in the treatment of PD. Trials on the effectiveness of MARI (moclobemide) do not show such clear results. Due to the potentially serious side effects and interactions with other drugs and dietary components, its use is recommended only when other drugs have failed^70,82.

When the response to the optimal dosage of one of the SSRIs is inadequate or if they are not well tolerated, the patient should switch to another SSRI. If there is no improvement after 8-12 weeks, consider using another drug with a different mechanism of action (NSRI, TAD, mirtazapine)^70,80.

Alprazolam, clonazepam, lorazepam, and diazepam have been proven to be effective in the treatment of PD, and the Canadian guide includes them as the second-tier pharmacological treatment^70,98. Alprazolam has been proven to reduce the frequency of panic attacks and symptoms of agoraphobia and anticipatory anxiety¹⁶⁶. The slow-release formula appears to have good initial speed of effect, with the advantage of longer duration of its therapeutic action¹⁷⁰. The short-term use of clonazepam at the start of treatment, along with SSRIs can result in faster response⁶¹. However, BDZs are associated with a wide spectrum of adverse effects both during and after treatment (dependence, withdrawal syndrome when stopped, and recurrence if treatment is discontinued). This effect of sudden withdrawal has been observed in patients with PD with alprazolam^82,98,171.

Since the long-term use of BDZs is associated with problems, their use is recommended for a limited time (short term), with the lowest possible dosage to reduce symptoms of PD, with the dosage reduced gradually. Short-term use is also recommended at any time to reduce agitation or acute or serious anxiety. The selected guides do not recommend long-term use, but if used, they add that usage must be supervised^70,80-82.

Existing evidence is insufficient to determine whether during pregnancy, the potential benefits of BDZs for the mother outweigh the possible risks to the fetus^142-144. To avoid the potential risk of congenital defects, the lowest effective dosage of BDZs should be used, with the shortest possible treatment duration, and as monotherapy. If higher concentrations are required, the daily dosage should be divided into two or three doses, always avoiding use during the first trimester¹⁴³. In advanced stages of pregnancy or during nursing, BDZs can cause adverse effects in neonatal infants (neonatal hypotonia, withdrawal syndrome, sedation, and hypothermia)^{82,87,146-149}.

The effectiveness of azapirones (buspirone) in the treatment of PD has not been clearly demonstrated so their use is not recommended^70,98,102.

Open studies suggest that atypical anti-psychotics olanzapine, quetiapine, and risperidone, added to anti-depressive treatment, may have some benefit in the treatment of refractory PD^70,155,172.

Other⁷⁰

1. Pindolol: beta-blocker and antagonist of the 1A seratoninergic receptor (5-HT 1A), added to fluoxetine appears to improve symptoms in treatment-resistant PD patients.
2. Gabapentine: anti-convulsive with a certain degree of effectiveness in gravely ill patients.
3. Sodium valproate (anti-convulsive) and slow-release bupropion (anti-depressant): these have demonstrated a certain degree of effectiveness in open trials.

Other agents such as tradozone (anti-depressant), propanolol (beta-blocker) and carbamazepine (anti-convulsive) have not been shown to be effective in the treatment of PD, so their use is not recommended.

In terms of pharmacological treatment to maintain benefits, SSRIs and TADs are the treatments of choice. In a study with a duration from 8 to 10 weeks, which included patient groups treated with a placebo, imipramine, or with fluoxetine, it was concluded that the patients treated with either imipramine or fluoxetine showed a statistically significant effect in the reduction of the number of spontaneous panic attacks, compared with the patients treated with the placebo¹⁷³.

A systematic review¹⁷⁴ that analyzes the data from nine RCTS with a sample size of 1,400 individuals, where the participants were patients were diagnosed with anxiety disorders, examines the effectiveness of CBT combined with medication (benzodiazepines, azaspirones, anti-depressants) in comparison with the effectiveness of each one of these therapies separately, measuring it with different scales (the scale of phobic avoidance, the WP2 anxiety inhibition scale, panic disorder severity scale – PDSS, and the Hamilton anxiety scale – HAM-A). It concludes that there are few studies that are methodologically suited for determining whether combined therapy is between than monotherapy, but those that directly compare combined therapy with pharmacological treatment show that combined treatment is better, which suggests that the use of a cycle of CBT should be considered for patients that have obtained partial response from medications after receiving pharmacotherapy alone. Also, with the exception of PD, adding drugs to CBT appears to not interfere negatively in the effects achieved over the long term with CBT alone.

In one RCTS¹⁷⁵ done in the context of Primary Care and in which psychologists provide CBT in the healthcare facility, CBT was combined with the use of diazepam. The CBT consisted of 7 sessions over a period of nine weeks. The patients received interventions such as cognitive therapy and progressive muscle relaxation, and took work home, working on the technique of exposure to situations and thoughts that generate anxiety. The evaluation of the effectiveness, using the Hamilton scale, continued until six months after the treatment. The study concluded that there is an advantage, in terms of severity and overall change in symptoms, of the combined treatment over the use of diazepam alone, but not over the use of CBT alone. The CBT either alone or in combination with a drug or placebo, showed the lowest incidence of referrals to psychologists and/or psychiatrists at a six-month follow-up.

The authors highlighted the need for more RCTSs that directly compare combined treatment with CBT and treatment with drugs. They also indicate that research should take into account patient preferences, long-term results, and the organizational coordination that the clinical practice reflects.

In our context, experiences have shown both a direct¹⁷⁶ and indirect^93,94, improvement in symptoms when group relaxation and cognitive interventions are added to pharmacological treatment. These interventions are carried out in the healthcare facility, mainly by nursing staff and/or social workers.

The first trials that examine the effectiveness of CBT combined with medication (benzodiazepines, azaspirones, anti-depressants) as compared to the effectiveness of each one of these therapies used separately show that the differences between these treatments are not very clear. Also, during follow-up, it was observed that combined therapy interfered with the maintenance of the benefits obtained long-term by BCT^177-179.