Adult female acne and associated risk factors: Results of a multicenter case-control study in Italy

Published:August 16, 2016DOI:https://doi.org/10.1016/j.jaad.2016.06.060

      Background

      The reasons for the appearance of acne in adulthood are largely unknown.

      Objective

      We explored the role of personal and environmental factors in adult female acne.

      Methods

      We conducted a multicenter case-control study in the outpatient departments of 12 Italian cities. Cases (n = 248) were consecutive women ≥25 years of age with newly diagnosed acne of any grade. Controls (n = 270) were females diagnosed with conditions other than acne.

      Results

      In multivariate analysis, a history of acne in parents (odds ratio [OR] = 3.02) or siblings (OR = 2.40), history of acne during adolescence (OR = 5.44), having no previous pregnancies (OR = 1.71), having hirsutism (OR = 3.50), being an office worker versus being unemployed or being a housewife (OR = 2.24), and having a high level of reported psychological stress (OR = 2.95) were all associated with acne. A low weekly intake of fruits or vegetables (OR = 2.33) and low consumption of fresh fish (OR = 2.76) were also associated with acne.

      Limitations

      We did not establish an onset date for acne. Some of our associations may reflect consequences of established acne.

      Conclusion

      Lifestyle factors may play an important role for acne development in adulthood, but their role should be further assessed in prospective studies.

      Key words

      • Several factors have been associated with adult acne.
      • Adult female acne was associated with a family history of acne, acne in adolescence, no previous pregnancies, and lifestyle factors (eg, diet, occupation, and stress).
      • Lifestyle factors should be given proper attention in the management of women with adult acne.

      Introduction

      Acne is one of the most common diseases in adolescence, affecting >80% of young people.
      • Krowchuk D.P.
      Managing adolescent acne: a guide for pediatricians.
      The disease usually resolves after 20 years of age, although it also affects 20% to 40% of adults. A study of people 40 to 49 years of age conducted in 1979 showed a prevalence of acne of about 3% in men and 11% to 12% in women, with a significant decrease after 45 years of age.
      • Cunliffe W.J.
      • Gould D.J.
      Prevalence of facial acne vulgaris in late adolescence and in adults.
      Over the last decades, the average age of people with acne has increased from 20.5 to 26.5 years.
      • Golden V.
      • Clark S.M.
      • Cunliffe W.J.
      Post-adolescent acne: a review of clinical features.
      In a 2001 study of 3305 women 25 to 40 years of age, the prevalence of acne was 41%,
      • Poli F.
      • Dreno B.
      • Verschoore M.
      An epidemiological study of acne in female adults: results of a survey conducted in France.
      and in a second study of 2895 women 10 to 70 years of age acne peaked during the teenage years, but >25% of women suffered from acne after 21 years of age, with a peak of 45% in women between 21 and 30 years of age.
      • Perkins A.C.
      • Maglione J.
      • Hillebrand G.G.
      • et al.
      Acne vulgaris in women: prevalence across the life span.
      When the disease is present after the age of 25 years in women, it is usually named adult female acne (AFA), which is comprised of 2 variants: “persistent” acne continues after adolescence, while “late onset” acne starts after 25 years of age.
      • Golden V.
      • Clark S.M.
      • Cunliffe W.J.
      Post-adolescent acne: a review of clinical features.
      The latter seems to occur less frequently than the former.
      • Preneau S.
      • Dreno B.
      Female acne - a different subtype of teenager acne?.
      The reasons for its increase in adulthood are unknown: endogenous and exogenous factors, such as smoking and psychological stress, can combine and contribute to its clinical expression.
      • Schmitt J.V.
      • Masuda P.Y.
      • Miot H.A.
      Acne in women: clinical patterns in different age-groups.
      • Schäfer T.
      • Nienhaus A.
      • Vieluf D.
      • et al.
      Epidemiology of acne in the general population: the risk of smoking.
      • Capitanio B.
      • Sinagra J.L.
      • Bordignon V.
      • et al.
      Underestimated clinical features of postadolescent acne.
      To analyze the roles played by different factors, including family history, smoking habits, occupation, comorbidities, psychological stress, and dietary factors, we conducted a case-control study of adult women who were clinically diagnosed with acne at several dermatologic outpatient clinics in Italy.

      Methods

      This multicenter case-control study assessed factors associated with AFA in a consecutive group of women ≥25 years of age who attended a dermatologic outpatient department in 1 of 12 Italian cities distributed across Italy. The patients were diagnosed with acne of any grade (ranging from mild to severe), as assessed by dermatologists during the visit, while the controls were women ≥25 years of age who visited a dermatologic outpatient clinic for a condition other than acne and who were not diagnosed with acne during the visit. All patients provided written informed consent before participating in the study. The protocol was reviewed and approved by the ethics committee at each participating center.
      Acne severity was defined according to a global score,
      • Di Landro A.
      • Cazzaniga S.
      • Parazzini F.
      • et al.
      Family history, body mass index, selected dietary factors, menstrual history and risk of moderate to severe acne in adolescents and young adults.
      using photographs limited to the face to help with the assessments. The following 4 categories were considered: no or minimal acne lesions, where only a few comedones are present; mild acne, where lesions include several noninflammatory comedones with <10 inflammatory lesions; moderate acne, with many comedones, papules, and pustules but no nodules; and severe acne, with inflammatory nodules in addition to papules and pustules. The “inflammatory” form was considered when the acne was dominated by inflammatory papules and pustules
      • Golden V.
      • Clark S.M.
      • Cunliffe W.J.
      Post-adolescent acne: a review of clinical features.
      ; this form was distinct from a “retentional” form, in which the comedones and microcysts predominated, with minimal inflammatory signs.
      A standardized questionnaire was administered by dermatologists during outpatient visits and was developed to collect general sociodemographic information (eg, age, sex, education, and occupation), personal habits (eg, smoking and alcohol consumption), anthropometric measurements (eg, height and weight), pregnancy, menstrual pattern and history, use of oral contraceptives, history of adolescence acne, history of acne in relatives, relevant comorbidities (eg, polycystic ovary syndrome [PCOS], hirsutism, type II diabetes, and thyroid disease), and stress level during last month, which was self-assessed according to a 5-point scale (ie, none, mild, moderate, high, and very high).
      A food frequency questionnaire was used to record information on the intake of selected food items, including the usual number of portions per week in the last month before the interview. Intakes of foods that were reported at least once a month but less than once a week were coded as 0.5 portions per week. The food groups investigated included whole, skim, or partially skim milk, other dairy products (eg, cheese and yogurt), starchy foods (eg, pasta, bread, and rice), fish, beef, chocolate, cakes and sweets, vegetables, and fruit.
      Characteristics of acne (eg, severity, location, and distribution) and drugs taken for the disease were also investigated.

       Statistical analysis

      For descriptive purposes, data were presented as means with standard deviations (SD) or as numbers with percentages for continuous and categorical variables, respectively. Univariate differences between the cases and controls (across different levels of variables) were assessed using the Mann–Whitney U test for continuous variables and the Pearson's chi-squared or Fisher's exact tests, as required, for categorical variables. For analytical purposes, continuous variables were categorized based on clinically meaningful values or tertile data distribution as cut-off thresholds. Food variables were categorized using the first tertile of distribution as the cut-off point. All variables with a P value < .15 in the univariate analysis were evaluated through 4-stage hierarchical multiple logistic regression, with a forward stepwise algorithm, to assess which variables were independent factors associated with AFA. In the first stage of the model, the patients' constitutional factors (ie, acne in first-degree relatives, including parents and siblings) were evaluated; in the second stage, other general variables, including occupation, pregnancy and menstrual history, and history of adolescence acne were assessed. In the next stage of the model, relevant comorbidities and stress level during the last month were considered; in the final stage, food consumption frequencies were evaluated. Variables selected at each stage were retained as fixed factors in the next step of the algorithm. Age was included as a general fixed adjustment factor in the models. The effects of the identified factors were presented as odds ratios (ORs), with 95% confidence intervals (CIs) and P values. Patients with missing data were excluded from the analysis. All tests were considered to be significant at P < .05. The analyses were performed using SPSS software (version 20.0; IBM Corporation, Armonk, NY).

      Results

       Demographics and clinical characteristics

      From February 2013 to January 2015, 518 women (248 cases and 270 controls) were included in the study. Table I shows the demographic distribution and general characteristics of the subjects according to their case or control status. Overall, the mean age was 32.2 ± 5.2 and 36.4 ± 7.0 years (mean ± SD) for cases and controls, respectively (P < .001), and the mean body mass index (BMI) was 22.5 ± 3.9 kg/m2. A total of 24.5% of the women were current smokers, and 62.3% were regular or occasional drinkers, with no significant differences between the cases and controls. A large proportion of the women reported attending upper secondary school or having a university degree (87.1%), and many (45.3%) worked as office workers; the differences between the cases and controls did not reach the level of significance (P = .10). A history of previous pregnancy was present in 39.1% of the cases compared to 62.2% of the controls (P < .001), and the age at menarche was slightly lower among the cases compared to the controls (P = .03).
      Table IDemographic distribution and general characteristics of the 518 subjects included in the study, both overall and according to their case-control status
      AFAP value
      Mann–Whitney U test and Pearson's chi-squared or Fisher's exact tests, as required, were used to assess case-control differences in continuous and categorical variables, respectively.
      No (n = 270)Yes (n = 248)Total (N = 518)
      n
      Numbers may not add up to the total because of missing data.
      %n
      Numbers may not add up to the total because of missing data.
      %n
      Numbers may not add up to the total because of missing data.
      %
      Age, y (mean ± SD)36.4 ± 7.032.2 ± 5.234.4 ± 6.6<.001
       25-295620.78634.714227.4
       30-3911542.614156.925649.4
       40+9936.7218.512023.2
      Body mass index, kg/m2 (mean ± SD)22.6 ± 3.922.4 ± 3.922.5 ± 3.9.48
       <20.06825.86526.313326.0
       20.0-24.914755.713755.528455.6
       ≥25.04918.64518.29418.4
      Smoking habits
       No/ex-smoker20676.318574.639175.5.65
       Current smoker6423.76325.412724.5
      Drinker
       No/ex-drinker10538.99036.419537.7.57
       Occasional/regular16561.115763.632262.3
      Educational attainment
       Lower secondary school4014.92610.76612.9.34
       Upper secondary school11844.011647.723445.8
       University/doctorate degree11041.010141.621141.3
      Present or last occupation
       None/housewife6825.54317.811121.9.10
       Student166.0218.7377.3
       Unskilled worker2810.52510.45310.4
       Office worker11041.212049.823045.3
       Manager/freelance4516.93213.37715.2
      History of pregnancy
       No10237.815160.925348.8<.001
       Yes (1+)16862.29739.126551.2
      Age at menarche, y (mean ± SD)12.6 ± 1.512.4 ± 1.512.5 ± 1.5.03
       <124018.65928.29923.3
       125827.05224.911025.9
       ≥1311754.49846.921550.7
      AFA, Adult female acne; SD, standard deviation.
      Mann–Whitney U test and Pearson's chi-squared or Fisher's exact tests, as required, were used to assess case-control differences in continuous and categorical variables, respectively.
      Numbers may not add up to the total because of missing data.
      Most of the cases had moderate (50.2%) or mild acne (42.0%), with only 19 patients (7.8%) showing signs of severe acne. Inflammatory acne was the most common clinical form (75.2%), and most of the subjects had facial lesions (96.8%), particularly on the cheeks (74.7%) and chin (71.3%), with an average number of 16.6 ± 11.3 lesions. In addition, 30.4% of the patients had lesions located on their trunks. The most common lesions observed were papules (35.4%), pustules (32.1%), and comedones (30.0%). The mean age of current acne onset was 30.5 ± 6.4 years. Controls were mainly visited for pigmentary lesion screening (36.3%), warts (5.6%), dermatitis/eczema (5.2%), and psoriasis (4.8%). A full list of diagnoses is provided in Supplemental Table I (available online at www.jaad.org).

       Medical histories

      The subjects' medical histories, overall and stratified by case or control status, are shown in Table II. A history of acne during adolescence was reported in 68.1% of the cases compared with 28.5% of the controls (P < .001). A family history of acne was present in 42.7% of the women, with a significant between-group difference (P < .001), although this finding was relevant only for first-degree relatives (ie, parents and siblings). Regarding the presence of comorbidities, a diagnosis of PCOS or hirsutism was reported by 21.8% of women, and it was significantly associated, at the univariate level, with an increased risk of AFA. A total of 22.3% of the women reported having irregular menstrual patterns, and 24.2% reported using oral contraceptives, without any significant differences between the cases and controls. A total of 27.1% of the subjects reported having a high or very high level of stress during the last month, with an overall significant difference between the 2 groups (P = .001).
      Table IIMedical history, both overall and according to case-control status
      AFAP value
      Pearson's chi-squared or Fisher's exact tests, as required, were used to assess case-control differences.
      NoYesTotal
      n
      Numbers may not add up to the total because of missing data.
      %n
      Numbers may not add up to the total because of missing data.
      %n
      Numbers may not add up to the total because of missing data.
      %
      History of adolescent acne
       No19371.57931.927252.5<.001
       Yes7728.516968.124647.5
      History of acne in relatives
       No17665.911848.029457.3<.001
       Yes9134.112852.021942.7
      Parents
       No23689.117571.741180.7<.001
       Yes2910.96928.39819.3
      Siblings
       No22283.817471.339677.8.001
       Yes4316.27028.711322.2
      Cousins
       No24893.623094.347893.9.75
       Yes176.4145.7316.1
      Uncles/aunts
       No25897.423696.749497.1.67
       Yes72.683.3152.9
      Comorbidities
       PCOS
      No22383.217972.840278.2.004
      Yes4516.86727.211221.8
       Hirsutism
      No25796.320482.646189.7<.001
      Yes103.74317.45310.3
       Type II diabetes
      No26799.624398.051098.8.11
      Yes10.452.061.2
       Thyroid disease
      No22884.421386.244185.3.57
      Yes4215.63413.87614.7
       Other diseases
      No20978.919078.539978.7.92
      Yes5621.15221.510821.3
      Menstrual pattern
      Postmenopausal women were excluded.
       Regular20278.018977.539177.7.89
       Irregular5722.05522.511222.3
      Use of oral contraceptives
       No20376.918574.638875.8.54
       Yes6123.16325.412424.2
      Stress during last month
       No/mild9034.05120.914127.7.001
       Moderate11744.211346.323045.2
       High/very high5821.98032.813827.1
      AFA, Adult female acne; PCOS, polycystic ovary syndrome.
      Pearson's chi-squared or Fisher's exact tests, as required, were used to assess case-control differences.
      Numbers may not add up to the total because of missing data.
      Postmenopausal women were excluded.

       Food intake

      Food intake frequency, overall and according to the case-control status, is shown in Table III. Regular milk consumption (>3 days/week) was recorded in 55.0% of the patients, without any significant difference in milk type between the cases and controls. In total, 51.1% of the women regularly ate dairy products, 83.8% consumed starchy foods, 39.3% ate cakes and sweets, 29.5% consumed chocolate, and 20.5% ate beef, without any significant between-group differences. The rate of women consuming vegetables and fruit or fish was significantly higher in the controls compared to the cases (P = .02 and P = .05, respectively).
      Table IIIFood intake frequency, both overall and according to case-control status
      AFAP value
      Pearson's chi-squared or Fisher's exact tests, as required, were used to assess case-control differences.
      NoYesTotal
      n
      Numbers may not add up to the total because of missing data.
      %n
      Numbers may not add up to the total because of missing data.
      %n
      Numbers may not add up to the total because of missing data.
      %
      Milk (days/week)
       ≤311442.211648.123045.0.18
       >315657.812551.928155.0
      Milk type
       Skimmed3417.93118.56518.2.96
       Partially skimmed11258.910059.521259.2
       Whole4423.23722.08122.6
      Dairy products (days/week)
       ≤312646.712451.525048.9.28
       >314453.311748.526151.1
      Starchy foods (days/week)
       ≤34115.24217.48316.2.49
       >322984.819982.642883.8
      Cakes and sweets (days/week)
       ≤316059.315062.231060.7.49
       >311040.79137.820139.3
      Chocolate (days/week)
       ≤318468.117673.036070.5.23
       >38631.96527.015129.5
      Vegetables and fruit (days/week)
       ≤33111.54619.17715.1.02
       >323988.519580.943484.9
      Fish (days/week)
       ≤322784.121790.044486.9.05
       >34315.92410.06713.1
      Beef (days/week)
       ≤321378.919380.140679.5.74
       >35721.14819.910520.5
      AFA, Adult female acne.
      Pearson's chi-squared or Fisher's exact tests, as required, were used to assess case-control differences.
      Numbers may not add up to the total because of missing data.

       Multivariable analysis

      The multivariable analysis results of the variables associated with AFA in the univariate analysis are shown in Table IV. After adjusting for age, a history of acne in first-degree relatives—either parents (OR = 3.02 [95% CI, 1.80-5.06]) or siblings (OR = 2.40 [95% CI, 1.46-3.94])—was significantly associated with AFA. The following variables were also associated with AFA: being an office worker (OR = 2.24 [95% CI, 1.24-4.06]) compared to being unemployed or a housewife, having a personal history of acne during adolescence (OR = 5.44 [95% CI, 3.43-8.61]), having no previous pregnancies (OR = 1.71 [95% CI, 1.06-2.78]), having hirsutism (OR = 3.50 [95% CI, 1.42-8.60]), and having a high or very high level of reported psychological stress during the last month (OR = 2.95 [95% CI, 1.57-5.53]) compared to having no stress or a mild level of stress. Regarding food consumption, a low weekly intake (≤3 days/week) of fruits and vegetables (OR = 2.33 [95% CI, 1.20-4.53]) and consuming fresh fish (OR = 2.76 [95% CI, 1.31-5.81]), were associated with AFA after controlling for all other variables in our multivariate analysis. No additional associations emerged.
      Table IVResults of multivariable analysis of the variables associated with adult female acne onset
      VariableOR
      Multistage hierarchical multiple logistic regression with forward stepwise selection algorithm; age was included as a fixed adjustment factor in the models.
      (95% CI)
      P value
      Constitutional factors
       History of acne in first-degree relatives
      Parents
      No1
      Yes3.02 (1.80-5.06)<.001
      Siblings
      No1
      Yes2.40 (1.46-3.94).001
      Other general factors
       Present or last occupation
      None/housewife1
      Student1.09 (0.44-2.73).85
      Unskilled worker1.46 (0.63-3.40).38
      Office worker2.24 (1.24-4.06).007
      Manager/freelance0.82 (0.39-1.72).60
       History of pregnancy
      No1.71 (1.06-2.78).02
      Yes (1+)1
       History of adolescent acne
      No1<.001
      Yes5.44 (3.43-8.61)
      Comorbidities and stress
       Hirsutism
      No1
      Yes3.50 (1.42-8.60).006
       Stress during last month
      No/mild1
      Moderate1.33 (0.76-2.30).32
      High/very high2.95 (1.57-5.53).001
       Food intake (days/week)
      Fruits/vegetables
      ≤32.33 (1.20-4.53).01
      >31
      Fresh fish
      ≤32.76 (1.31-5.81).008
      >31
      CI, Confidence interval; OR, odds ratio.
      Multistage hierarchical multiple logistic regression with forward stepwise selection algorithm; age was included as a fixed adjustment factor in the models.

      Discussion

      Few formal epidemiologic studies have evaluated risk factors for the presence of AFA. Our case-control study indicated that having a personal history of acne in adolescence, a family history of acne in first-degree relatives, no previous pregnancies, having hirsutism, working as an office worker, reporting a higher level of psychological stress, and having some dietetic factors, including a low consumption of vegetables or fruit and fish, were all associated with AFA. We did not establish an onset date for acne, and our associations may not reflect a causative role but may, at least in part, reflect shared risk factors or consequences of established acne. The small proportion of patients (7%) with severe acne was consistent with previous results reported in the literature,
      • Preneau S.
      • Dreno B.
      Female acne - a different subtype of teenager acne?.
      which often define AFA as a mild to moderate condition. A total of 75.2% of the patients had the inflammatory form of the disease rather than the retentional form. This finding aligned with data previously reported by Dréno et al.
      • Dréno B.
      • Thiboutot D.
      • Layton A.M.
      • et al.
      Large-scale international study enhances understanding of an emerging acne population: adult females.
      Our data also confirmed that the face is the predominant area affected by AFA, with lesions mainly localized at the cheeks and chin, which indicates the propensity of this clinical variant to be localized in the middle and inferior areas of the face.
      • Capitanio B.
      • Sinagra J.L.
      • Bordignon V.
      • et al.
      Underestimated clinical features of postadolescent acne.
      Truncal involvement was present in 30% of the cases. A history of acne in adolescence was reported by 68% of patients, indicating the predominance of a “persistent” AFA subtype compared to a “late onset” subtype, which aligns with the observations of other studies.
      • Dréno B.
      • Thiboutot D.
      • Layton A.M.
      • et al.
      Large-scale international study enhances understanding of an emerging acne population: adult females.
      The mean age of the cases (32.2 years) was similar to that in previous reports,
      • Shaw J.C.
      • White L.E.
      Persistent acne in adult women.
      • Dumont-Wallon G.
      • Dreno B.
      Specificity of acne in women older than 25 years.
      • Williams C.
      • Layton A.M.
      Persistent acne in women: implications for the patient and for therapy.
      confirming that AFA is primarily a young adult disease.
      The association of acne with a family history of the disease, which we documented, confirms data from other studies that also indicated a stronger association with a history in parents compared to siblings.
      • Golden V.
      • Clark S.M.
      • Cunliffe W.J.
      Post-adolescent acne: a review of clinical features.
      • Di Landro A.
      • Cazzaniga S.
      • Parazzini F.
      • et al.
      Family history, body mass index, selected dietary factors, menstrual history and risk of moderate to severe acne in adolescents and young adults.
      • Dréno B.
      • Thiboutot D.
      • Layton A.M.
      • et al.
      Large-scale international study enhances understanding of an emerging acne population: adult females.
      • Dumont-Wallon G.
      • Dreno B.
      Specificity of acne in women older than 25 years.
      • Knaggs H.E.
      • Wood E.J.
      • Rizer R.L.
      • et al.
      Post-adolescent acne.
      The association of AFA with hirsutism and having no previous pregnancies seems to support a connection between adult acne and hormonal factors, such as hyperandrogenism, even if the association between AFA and a history of PCOS was not confirmed in our multivariable analysis, possibly because of a limited statistical power to document such an association. We did not collect blood samples to assess hormone levels. Note that other studies have shown a weak association between AFA and underlying androgenic abnormalities.
      • Dréno B.
      • Thiboutot D.
      • Layton A.M.
      • et al.
      Large-scale international study enhances understanding of an emerging acne population: adult females.
      In our AFA study, we did not confirm the association between acne and BMI, which we documented in adolescent acne,
      • Di Landro A.
      • Cazzaniga S.
      • Parazzini F.
      • et al.
      Family history, body mass index, selected dietary factors, menstrual history and risk of moderate to severe acne in adolescents and young adults.
      where the association was much stronger in men compared with women. Interestingly, male (but not female) obesity is associated with hyperlipidemia, and high lipid levels have been associated with severe acne.
      • Arora M.K.
      • Seth S.
      • Dayal S.
      The relationship of lipid profile and menstrual cycle with acne vulgaris.
      Another debated association is with tobacco smoking, which was not documented in our AFA study, contrary to other experiences in which the disease, especially the comedonal noninflammatory variant, was strongly correlated with tobacco; one study referred to the condition as “smoker's acne.”
      • Capitanio B.
      • Sinagra J.L.
      • Bordignon V.
      • et al.
      Underestimated clinical features of postadolescent acne.
      The choice of proper controls might explain the different results obtained.
      In a previous case-control study, we highlighted the importance of dietary factors in acne occurring in adolescents. Acne was associated with a high consumption of milk and skimmed milk, and it was inversely correlated with the intake of fish.
      • Di Landro A.
      • Cazzaniga S.
      • Parazzini F.
      • et al.
      Family history, body mass index, selected dietary factors, menstrual history and risk of moderate to severe acne in adolescents and young adults.
      In the present study, we did not confirm the association with milk, although a regular intake of fish and fruits or vegetables were protective factors for the disease, possibly because of the antioxidant properties of these foods.
      • Skroza N.
      • Tolino E.
      • Semyonov L.
      • et al.
      Mediterranean diet and familial dysmetabolism as factors influencing the development of acne.
      These findings, together with the lack of an association with BMI, seem to indicate a different pathogenesis of AFA compared to adolescent acne, with a lower impact of metabolic factors.
      High or very high levels of psychological stress during the month preceding the clinical examination were also associated with AFA. This association is intriguing. Information on stress was self-reported by the patients and influenced by their perceptions. More convincing evidence could be derived from prospective studies assessing stressful events at several time points in relation with acne. Recent studies have shown that adult patients with acne are characterized by both common psychiatric conditions and a tendency to somatization.
      • Gül A.İ.
      • Çölgeçen E.
      Personality traits and common psychiatric conditions in adult patients with acne vulgaris.
      A high level of stress was also observed by Albuquerque et al,
      • Albuquerque R.G.
      • Rocha M.A.
      • Bagatin E.
      • et al.
      Could adult female acne be associated with modern life?.
      who showed that women with AFA are more prone to suffer from psychiatric conditions (eg, somatization, depression, and anxiety),
      • Albuquerque R.G.
      • Rocha M.A.
      • Bagatin E.
      • et al.
      Could adult female acne be associated with modern life?.
      for which a multidisciplinary approach would be required. The prevalence of active working women from 25 to 39 years of age was particularly high, and we observed an association between AFA and office-related work, which is difficult to explain and possibly related to the work environment (eg lighting, air conditioning, etc) or psychological factors.
      Our study had some limitations. We did not establish an onset date for acne, and some of our associations may reflect consequences of established acne. The age of patients with AFA was significantly lower than those without such a diagnosis, but we made adjustment by age in the multivariable analysis. Controls were recruited for a variety of clinical conditions unrelated to AFA. They were matched with cases on the setting of observation (ie, outpatient service), to control for factors related with seeking medical advice and the degree of attention for their health conditions.
      In conclusion, our case-control study is one of the few studies adopting a formal epidemiologic methodology to assess the risk factors for AFA in women ≥25 years of age. Our study indicated that having a history of acne in adolescence, a history of acne in first-degree relatives, no previous pregnancies, having hirsutism, working as an office worker, and reporting a higher level of psychological stress—along with some dietetic factors, including low consumption of vegetables or fruit and fish—were all associated with AFA. Some of our associations could be viewed as explorative and therefore require further confirmation by other studies.

      Appendix

      Supplemental Table IList of diagnoses in control patients
      Diagnosis, n (%)Controls (N = 270)
      Pigmentary lesion screening98 (36.3)
      Warts15 (5.6)
      Unspecified dermatitis/eczema14 (5.2)
      Psoriasis13 (4.8)
      Basal cell carcinoma10 (3.7)
      Androgenetic alopecia9 (3.3)
      Allergic contact dermatitis9 (3.3)
      Mycosis/onychomycosis8 (3.0)
      Seborrheic dermatitis7 (2.6)
      Atopic dermatitis6 (2.2)
      Seborrheic keratosis4 (1.5)
      Sebaceous cyst4 (1.5)
      Skin biopsy/surgery3 (1.1)
      Melasma3 (1.1)
      Vitiligo3 (1.1)
      Vaginitis3 (1.1)
      Acne rosacea2 (0.7)
      Angioma2 (0.7)
      Condylomata2 (0.7)
      Rash2 (0.7)
      Lentigo2 (0.7)
      Papular lesions2 (0.7)
      Lichen planopilaris2 (0.7)
      Chronic urticaria2 (0.7)
      Pityriasis versicolor2 (0.7)
      Telogen effluvium2 (0.7)
      Dermatofibroma2 (0.7)
      Actinic keratosis1 (0.4)
      Pendulous fibroids1 (0.4)
      Folliculitis1 (0.4)
      Hidradenitis suppurativa1 (0.4)
      Syphilis1 (0.4)
      Melanoma1 (0.4)
      Molluscum contagiosum1 (0.4)
      Pyoderma1 (0.4)
      Pyoderma gangrenousum1 (0.4)
      Pemphigus1 (0.4)
      Pityriasis rosea1 (0.4)
      Pruritus1 (0.4)
      Purpura1 (0.4)
      Plantar callus1 (0.4)
      Tinea pedis1 (0.4)
      Burn1 (0.4)
      Other/unknown23 (8.5)

      References

        • Krowchuk D.P.
        Managing adolescent acne: a guide for pediatricians.
        Pediatr Rev. 2005; 26: 250-261
        • Cunliffe W.J.
        • Gould D.J.
        Prevalence of facial acne vulgaris in late adolescence and in adults.
        Br Med J. 1979; 1: 1109-1110
        • Golden V.
        • Clark S.M.
        • Cunliffe W.J.
        Post-adolescent acne: a review of clinical features.
        Br J Dermatol. 1997; 136: 66-70
        • Poli F.
        • Dreno B.
        • Verschoore M.
        An epidemiological study of acne in female adults: results of a survey conducted in France.
        J Eur Acad Dermatol Venereol. 2001; 15: 541-545
        • Perkins A.C.
        • Maglione J.
        • Hillebrand G.G.
        • et al.
        Acne vulgaris in women: prevalence across the life span.
        J Womens Health. 2012; 21: 223-230
        • Preneau S.
        • Dreno B.
        Female acne - a different subtype of teenager acne?.
        J Eur Acad Dermatol Venereol. 2012; 26: 277-282
        • Schmitt J.V.
        • Masuda P.Y.
        • Miot H.A.
        Acne in women: clinical patterns in different age-groups.
        An Bras Dermatol. 2009; 84: 349-354
        • Schäfer T.
        • Nienhaus A.
        • Vieluf D.
        • et al.
        Epidemiology of acne in the general population: the risk of smoking.
        Br J Dermatol. 2001; 145: 100-104
        • Capitanio B.
        • Sinagra J.L.
        • Bordignon V.
        • et al.
        Underestimated clinical features of postadolescent acne.
        J Am Acad Dermatol. 2010; 63: 782-788
        • Di Landro A.
        • Cazzaniga S.
        • Parazzini F.
        • et al.
        Family history, body mass index, selected dietary factors, menstrual history and risk of moderate to severe acne in adolescents and young adults.
        J Am Acad Dermatol. 2012; 67: 1129-1135
        • Dréno B.
        • Thiboutot D.
        • Layton A.M.
        • et al.
        Large-scale international study enhances understanding of an emerging acne population: adult females.
        J Eur Acad Dermatol Venereol. 2015; 29: 1096-1106
        • Shaw J.C.
        • White L.E.
        Persistent acne in adult women.
        Arch Dermatol. 2001; 137: 1252-1253
        • Dumont-Wallon G.
        • Dreno B.
        Specificity of acne in women older than 25 years.
        Presse Med. 2008; 37: 585-591
        • Williams C.
        • Layton A.M.
        Persistent acne in women: implications for the patient and for therapy.
        Am J Clin Dermatol. 2006; 7: 281-290
        • Knaggs H.E.
        • Wood E.J.
        • Rizer R.L.
        • et al.
        Post-adolescent acne.
        Int J Cosmet Sci. 2004; 26: 129-138
        • Arora M.K.
        • Seth S.
        • Dayal S.
        The relationship of lipid profile and menstrual cycle with acne vulgaris.
        Clin Biochem. 2010; 43: 1415-1420
        • Skroza N.
        • Tolino E.
        • Semyonov L.
        • et al.
        Mediterranean diet and familial dysmetabolism as factors influencing the development of acne.
        Scand J Public Health. 2012; 40: 466-474
        • Gül A.İ.
        • Çölgeçen E.
        Personality traits and common psychiatric conditions in adult patients with acne vulgaris.
        Ann Dermatol. 2015; 27: 48-52
        • Albuquerque R.G.
        • Rocha M.A.
        • Bagatin E.
        • et al.
        Could adult female acne be associated with modern life?.
        Arch Dermatol Res. 2014; 306: 683-688