Skepticism and pharmacophobia toward medication may negatively impact adherence to psychiatric medications: a comparison among outpatient samples recruited in Spain, Argentina, and Venezuela

Figures & data

Table 1 Sociodemographic and clinical characteristics of the samples studied (1,291 psychiatric patients, 2,308 psychiatric drugs used)

	Global N=1,291	Spain n=588	Argentina n=508	Venezuela n=195	p-value
Age, years	44.1±14.2	45.6±13.1	42.1±15.5	44.5±13.3	0.000a
18–35	31%	24%	40%	29%
35–50	34%	39%	28%	35%
50–65	28%	30%	24%	31%
>65	7%	7%	8%	5%
Sex					0.000b
Male	42%	47%	34%	47%
Female	58%	53%	66%	53%
Educational level					0.000b
Can read and write	1%	3%	0%	1%
Primary school	23%	27%	19%	19%
Secondary school	44%	44%	48%	34%
University	32%	26%	36%	47%
Diagnosis					0.000b
Schizophrenia	18%	23%	6%	31%
Bipolar disorder	11%	7%	12%	25%
Depressive disorder	41%	38%	54%	15%
Anxiety disorders	24%	29%	20%	20%
Personality disorders	6%	3%	8%	9%
Number of drugs	2,308	1,115	874	317
Psychotropic drugs	1.8±0.8	1.9±1.1	1.7±0.8	1.6±0.8	0.001b
One drug	49%	49%	46%	55%
Two drugs	33%	26%	41%	29%
Three drugs	12%	14%	9%	15%
Four drugs	4%	7%	3%	1%
Five or more drugs	2%	4%	1%	0%
Type of drug					0.000b
Antipsychotics	22%	19%	18%	46%
Mood stabilizers	13%	11%	14%	13%
Antidepressants	37%	35%	49%	12%
Antianxiety	28%	35%	19%	29%
Sidorkiewicz adherence tool					0.000b
High adherence	44%	56%	48%	49%
Good adherence	7%	8%	4%	7%
Moderate adherence	15%	6%	13%	11%
Poor adherence	12%	16%	6%	13%
Very poor adherence	6%	6%	7%	6%
Discontinuation	16%	18%	22%	14%
Adherence					0.145b
Yes	67%	66%	69%	64%
No	33%	34%	31%	36%
DAI-10	3.6±3.8	2.5±4.0	4.7±3.0	3.6±3.8	0.000b
Pharmacophobic (range −10 to 0)	23%	35%	9%	23%
Pharmacophilic (range >0 to 10)	77%	65%	91%	77%
BMQ-Necessity (range 1–5)	3.5±0.9	3.5±1.0	3.6±0.7	3.4±1.0	0.002a
BMQ-Concern (range 1–5)	3.0±1.0	2.8±1.1	3.3±0.7	2.8±1.0	0.000a
Attitudinal group					0.043b
Skeptical (low necessity-high concern)	27%	30%	23%	25%
Ambivalent (high necessity-high concern)	27%	25%	30%	28%
Indifferent (low necessity-low concern)	19%	18%	19%	19%
Accepting (high necessity-low concern)	27%	27%	28%	28%

Notes:

a Student’s t-test;

b chi-square. Data presented as mean ± SD or %.

Abbreviations: DAI-10, Drug Attitude Inventory – 10 Item; BMQ, Beliefs about Medicines Questionnaire.

Table 2 Analyses of the Necessity–Concerns Framework variables by type of psychoactive drug and country of origin

	BMQ-specific		Attitudinal group
	Necessity	Concern	Skeptical (%)	Ambivalent (%)	Indifferent (%)	Accepting (%)
Spain
Antipsychotics	3.71±1.0	2.61±1.1	22	23	15	40
Mood stabilizers	3.82±0.92	2.74±1.1	19	32	14	35
Antidepressants	3.37±0.98	2.76±1.0	32	21	23	24
Antianxiety	3.38±1.1	2.93±1.0	35	28	18	20
Total	3.49±1.0 p<0.000	2.79±1.1 p=0.009	30 χ^2=54.500, p<0.000	25	19	27
Argentina
Antipsychotics	3.56±0.79	3.47±0.85	28	35	14	24
Mood stabilizers	3.63±0.77	3.25±0.85	29	30	15	27
Antidepressants	3.63±0.66	3.24±0.71	18	30	21	31
Antianxiety	3.56±0.74	3.32±0.65	29	28	21.0	23
Total	3.60±0.7 p=0.756	3.30±0.7 p=0.021	24 χ^2=23.173, p=0.026	30	19	28
Venezuela
Antipsychotics	3.61±1.1	2.59±1.0	19	25	16	40
Mood stabilizers	3.35±1.0	2.59±0.90	19	30	23	28
Antidepressants	2.92±1.2	2.75±1.0	21	26	40	13
Antianxiety	3.28±1.1	3.04±0.98	37	32	14	18
Total	3.40±1.1 p=0.006	2.73±1.0 p=0.023	25 χ^2=35.185, p<0.000	28	19	29

Note: Data presented as mean ± SD or %.

Abbreviation: BMQ, Beliefs about Medicines Questionnaire.

Table 3 Analyses of the DAI-10 and BMQ according to self-reported adherence and by country

	Global	Non-adherent	Adherent	χ^2
Spain
DAI-10 score	2.5±4.0	1.7±4.2	3.2±3.6	0.000
Pharmacophobic	35%	41%	28%	17.466, p<0.000
Pharmacophilic	65%	59%	72%
BMQ-Necessity	3.5±1.0	3.2±110	3.7±1.0	0.000
BMQ-Concern	2.8±1.1	3.0±1.1	2.7±1.0	0.000
Skeptical	29%	38%	25%	22.375, p<0.000
Ambivalent	25%	23%	26%
Indifferent	19%	19%	19%
Accepting	27%	21%	30%
Argentina
DAI-10 score	4.7±3.0	3.9±4.1	5.2±2.3	0.000
Pharmacophobic	9%	15%	5%	24.714, p<0.000
Pharmacophilic	91%	85%	95%
BMQ-Necessity	3.6±0.7	3.2±0.7	3.7±0.7	0.000
BMQ-Concern	3.3±0.7	3.6±0.8	3.2±0.7	0.000
Skeptical	23%	46%	13%	118.712, p<0.000
Ambivalent	30%	24%	33%
Indifferent	19%	17%	20%
Accepting	28%	13%	34%
Venezuela
DAI-10 score	3.6±3.8	2.3±4.1	4.5±3.5	0.000
Pharmacophobic	23%	35%	15%	15.962, p<0.000
Pharmacophilic	77%	65%	85%
BMQ-Necessity	3.4±1.0	3.1±1.1	3.5±1.0	0.001
BMQ-Concern	2.8±1.0	2.9±1.0	2.6±1.0	0.017
Skeptical	25%	34%	20%	11.309, p<0.010
Ambivalent	28%	26%	29%
Indifferent	19%	20%	18%
Accepting	28%	20%	33%

Note: Data presented as mean ± SD or %.

Abbreviations: BMQ, Beliefs about Medicines Questionnaire; DAI-10, Drug Attitude Inventory – 10 Item.

Table 4 Binary logistic regression analyses in the global sample and by country (variables entered: gender, age, pharmacophobia, and skepticism)a

Variable	Global sample	Spain	Argentina	Venezuela
Sexb
OR (CI)	1.35 (1.12–1.61)	1.17 (0.91–1.51)	1.45 (1.04–2.01)	1.90 (1.15–3.14)
p-value	0.001	0.22	0.03	0.012
χ^2	0.30/0.1/10.12/1	0.15/0.13/1.51/1	0.37/0.17/4.81/1	0.65/0.26/6.38/1
Age
OR (CI)	1.02 (1.01–1.03)	1.01 (1.00–1.02)	1.02 (1.01–1.03)	1.03 (1.01–1.04)
p-value	<0.001	0.02	<0.001	0.01
χ^2	0.02/0.003/31.4371	0.01/0.005/5.26/1	0.02/0.005/17.75/1	0.02/0.01/6.58/1
Pharmacophobiac
OR (CI)	1.66 (1.34–2.06)	1.52 (1.16–1.99)	2.17 (1.27–3.71)	2.61 (1.43–4.76)
p-value	<0.001	0.003	0.005	0.002
χ^2	0.30/0.09/10.1/1	0.42/0.14/9.03/1	0.77/0.27/7.95/1	0.96/0.30/9.81/1
Skepticismc
OR (CI)	2.22 (1.82–2.71)	1.56 (1.18–2.06)	4.39 (3.10–6.22)	1.40 (0.79–2.50)
p-value	<0.001	0.002	<0.001	0.25
χ^2	0.80/0.10/60.73/1d	0.44/0.14/9.93/1d	1.48/0.18/69.8/1d	0.34/0.29/1.33/1d
Model summary
2 Log likelihood	2,747.989	1,393.193	939.572	376.486
Cox and Snell R^2	0.072	1.032	0.145	0.101
Nagelkerke R^2	0.100	1.044	0.205	0.139
Hosmer and Lemeshow	0.349	0.206	0.002	0.340

Notes:

a The regression models presented are highly significant (p<0.001 in the Omnibus Test of the four models), reaching moderate degrees of adjustment and correctly classifying 92% of the adherence cases (Spain: 95%; Argentina: 90%; Venezuela: 90%), 70% of the total (Spain: 66%; Argentina: 74%; Venezuela: 70%), and only 23% of the cases with non-adherence (Spain: 10%; Argentina: 39%; Venezuela: 33%). Neither educational level nor diagnoses were included since they did not register significant differences in univariante analysis;

b 0= male and 1= female;

c 0= no and 1= yes;

d χ² is a chi-square test, which is summarized by using four statistical parameters: beta/standard error/Wald/degrees of freedom.

Abbreviations: CI, confidence interval; OR, odds ratio.

Figure 1 CHAID diagram of the tree with five predictors (country, sex, age, pharmacophobia, and skepticism) and adherence to prescribed treatment as the outcome variable in the global sample of psychiatric outpatients.

Abbreviations: Adj, adjusted; CHAID, chi-squared automatic interaction detector.

Table 5 Summary of results from CHAID decision trees for each country with their adherence classification

Results	Spain	Argentina	Venezuela
Percent correct
Non-adherent	23%	46%	35%
Adherent	90%	87%	85%
Overall	67%	74%	67%
Highest adherencea	70% in non-skeptical	79% in non-skeptical	70% in non-pharmacophobic
Highest non-adherencea	56% in skeptical and pharmacophobic	60% in skeptical	56% in pharmacophobic
Risk estimate	0.330	0.258	0.330
Standard error	0.014	0.015	0.026

Notes:

a Percentages in these two rows are inverse (eg, in Spain the group with the highest adherence was the non-skeptical patient group, including 70% who were adherent and 30% non-adherent; the group with the highest non-adherence was the skeptical and pharmacophobic patient group, including 56% who were non-adherent and 44% adherent).

Abbreviation: CHAID, chi-squared automatic interaction detector.

Table

• A great majority of psychiatric outpatients (>64%), independent of their country of origin, expressed a positive attitude toward their prescribed psychotropic medications, according to scores registered by the DAI-10.

• The Argentinian patients were those who self-reported a greater necessity for their prescribed psychotropic drugs as well as a greater concern about their use, while the Venezuelan patients self-reported the least necessity and concern about their possible side effects.

• The Spanish and Venezuelan patients registered significant differences on the scales of necessity and concern regarding the various psychotropic drugs consumed.

• For Spanish patients, the higher necessity drugs were self-reported to be mood stabilizers, while in Venezuelan patients the higher necessity drugs were antipsychotics. Both countries shared higher concern about benzodiazepines.

• The Argentine patients did not register significant differences in their perception of the necessity of the various psychotropic drugs, although they registered differences in their concerns about adverse effects.

• The greatest concern of the Argentinians was with antipsychotic drugs.

• The Spanish patients were the most skeptical about their prescribed treatment while the Venezuelans were those who most accepted their prescribed treatment.

• Benzodiazepines generated the greatest skepticism in Spanish and Venezuelans, while for the Argentineans, the mood stabilizer drugs registered greater skepticism.

• On the other hand, antipsychotics were the drugs most accepted by Spanish and Venezuelans, while for the Argentines it was antidepressants.

Table

• A look at the most crucial percentage of tables relevant for the planned multivariate analyses indicated:

○ An average lack of adherence of 33% (range 31%–36%)

○ An average pharmacophobia of 23% (range 9%–35%)

○ An average skeptical group of 27% (range 23%–30%)

• Regarding patient nationalities:

○ Venezuelans appear to be relatively similar to the Spaniards from the Canary Islands

○ Argentinians appear much more pharmacophilic than the other two countries with a very small percentage of patients (9%) classified as pharmacophobic