Epidemiology and Healthcare of Juvenile and Late-Onset Acne: Claims Data Analysis

Abstract

Introduction

Acne is the most frequent chronic inflammatory skin condition in adolescence but occurs also in later age. Our study aimed to characterise the epidemiology, geographical distribution, comorbidity and healthcare of acne juvenilis (AJ) and acne tarda (AT).

Methods

Statutory health insurance (SHI) data from 2016 to 2020 were analysed. Prevalence rates, including geographical distribution, comorbidities and drug use by specialists group, were measured.

Results

In 2020, the prevalence of acne among adults was 1.50% (AT > 25 years) and among adolescents was 3.88% (AT ≤ 25 years). The highest prevalence (13.02%) was observed at the age of 17 years. Sex differences were higher in AT (73.80% in women) than in AJ (64.55% in women). Compared with non-affected persons, individuals with acne – in particular with AT – showed significantly higher rates of skin-related comorbidities, including folliculitis (rate ratio (RR) 8.89), pyoderma (RR 7.27) and rosacea (RR 5.53), and non-skin-related comorbidities, such as ovarian dysfunction (RR 2.36), rhinitis allergica (RR 1.84) and Crohn’s disease (RR 1.79). Preferred systemic therapeutics prescribed were anti-infectives in AT (46.86%) and retinoids in AJ (52.35%). In the majority of cases, dermatologists were involved in the treatment of acne (AT 65.77%; AJ 76.27%). The most commonly prescribed topical agents were adapalene with benzoyl peroxide (AT 87.72%; AJ 85.75%), while systemic isotretinoin (AT 81.20%; AJ 90.06%) was also a frequently used drug. General practitioners were more likely to prescribe anti-infectives, especially topical antibiotics such as chlortetracycline (AT 52.38%; AJ 44.44%) and systemic antibiotics, especially tetracycline (AT 58.33%; AJ 67.50%).

Conclusion

Acne affects a relevant proportion of the German population not only in adolescence, and management of this inflammatory skin disease does not naturally follow medical guidelines or specialist recommendations. These findings emphasise the importance of specialised care and comprehensive therapeutic management that should also consider the exploration of comorbidities.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13555-024-01284-z.

Plain Language Summary

We performed an analysis on the frequencies of illness and concomitant diseases of acne in Germany based on health insurance data. Furthermore, the treatment of acne in regard to drug choices and prescribers was characterised. In 2020, about 2 million people in Germany had acne. The mean age was 31 years, and women were more often affected (68.58%). The highest proportion of acne was observed at the age of 17 years (13.02%). Overall, adolescent acne (younger than 25 years) was more frequent than adult acne (older than 25 years). People with acne, especially with adult acne, had higher rates of concomitant diseases (compared with non-affected persons, both related to the skin and beyond). About a third of people with acne were not treated by a dermatologist, and the majority of these patients were not treated according to guidelines. In conclusion, acne and related concomitant diseases affect a relevant proportion of adolescents and adults in Germany. The high prevalence of adult acne suggests a special focus also on this age group, including exploration of hormonal and psychological disorders. Early and effective treatment, adapted to the severity of the acne, can prevent complications such as scarring.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13555-024-01284-z.

Key Summary Points

Why carry out this study?

In addition to its occurrence in adolescence, acne is also common in adults, predominantly in women.

Our study aimed to characterise the epidemiology, geographical distribution, comorbidity and healthcare of acne juvenilis (AJ) and acne tarda (AT).

What was learned from the study?

Marked regional variations in the prevalence of AJ and AT were observed across Germany.

Individuals in all age groups but particularly with AT showed significantly higher rates of comorbidities of the skin and beyond.

Treatment without a dermatologist’s involvement was associated with a lower likelihood of guideline-concordant treatment.

Introduction

Acne is a chronic inflammatory skin condition with a postulated global point prevalence of 9.4% [1]. The estimated lifetime prevalence was 73.3% [2]. It occurs in up to 57.8% of adolescents up to the age of 24 years [3]; this type called acne juvenilis (AJ). However, acne can persist into adulthood or set-up de novo in older persons. Such subtypes have been called adult acne [4], late-onset acne [5], or acne tarda (AT) [6]. Estimates on the annual prevalence of acne vulgaris in the population mostly range from 3% to 4%, depending on the study methodology [7, 8]. The prevalence of acne also appears to vary by skin type and ethnic group [1, 9–11]. Adolescent acne is induced by hormonal factors triggering an increased sebum production and follicular hyperkeratinisation as a result of pubertal maturation [7]. In adults, pathological hormonal fluctuations or hypersensitivity to normal androgen levels might be the reason for acne [12–14].

Quality of life in people with acne is especially affected by the visibility of inflammatory and non-inflammatory lesions and the associated psychosocial distress [15, 16]. The residual scarring, which can lead to lifelong disfigurement, is also a particular burden. People with acne have a higher risk of cutaneous (rosacea and folliculitis) and systemic comorbidities (e.g. hyperlipidemia, diabetes and cardiovascular disease) [8, 17–21]. To our knowledge, there are only a few studies on the epidemiology and healthcare for acne in Germany and abroad [22–25]. In particular, a few population-based studies have been published that focus on the prevalence of acne and its comorbidities in adulthood and prescriber coverage. Nevertheless, no studies have been conducted to examine the impact of geographical location or the quality of care provided by the practitioner.

The aim of our study was to utilise population-based claims data to characterise the epidemiology, overall and by federal state; comorbidity; and drug use by specialist group for AT and AJ in Germany.

Methods

Study Design and Data Source

Statutory health insurance (SHI) is an essential part of the German healthcare system: about 89% of the German population (about 72 million) is insured by one of the 95 SHI funds (in 2024). The remaining 11% are privately insured [26]. The DAK-Gesundheit is a large and nationwide operating health insurance company, with 5.6 million members in 2019. The SHI system in Germany is based on the principle of solidarity. The financial capacity of members determines the extent of their contributions, while the healthcare services provided are uniform and accessible to all, regardless of age [27].

A claims data analysis was conducted on the basis of anonymised routine data of the German nationwide SHI company DAK-Gesundheit. The evaluation was carried out with a random 40% sample of all insurees (insured persons including family insured persons, N = 2,885,984; 56.80% women; average age 49.11 years) who were enlisted for at least 1 day between 1 January 2016 and 31 December 2020.

Study Population

The study population consists of persons insured on the 31 December of the year under observation. The identification of prevalent insured persons with acne [International Statistical Classification of Diseases and Related Health Problems (ICD) L70] in Germany was based on at least one relevant inpatient diagnosis (main or secondary diagnosis; ICD-10 GM: L70) or at least one confirmed outpatient diagnosis in the prevalence year (2016–2020).

Adult acne (AT) was defined as (a) occurring at the age of > 25 years in accordance with the World Health Organization (WHO) term “young people” (age between 10 and 24 years) and (b) at the age of > 30 years to compare with previous publications [28].

An a priori list of comorbid conditions of interest was defined on the basis of a literature review and clinical experience (Table S1 in the electronic supplementary material). Diseases in insured persons with acne in the prevalence year 2020 were considered as a comorbidity if at least one confirmed diagnosis was documented in an outpatient setting or at least one main or secondary inpatient diagnosis was coded. The comorbidity of persons with acne was compared with insurees without acne (no ICD-10-GM L70 diagnosis) in 2020. To exclude historical diagnoses and thus deadweight effects, the comorbidity of persons with acne treated with a relevant medication was also included as a sensitivity test (Table S2 in the electronic supplementary material).

Defined relevant drug prescriptions were included using the Anatomical Therapeutic Chemical (ATC) code (Table S2 in the electronic supplementary material) if an acne diagnosis was coded in the same quarter or in the quarter before the drug prescription. In addition, the drugs were presented according to specialist groups. The treating outpatient specialist groups were determined quarterly on the basis of specialist group codes. The specialist group codes were grouped as follows (Kassenärztliche Bundesvereinigung [National Association of Statutory Health Insurance Physicians], 2016–2020): General practitioner 01–03, Dermatologist 21, Internist 23–33, Paediatrician 34–47 and Gynaecologist 15–18.

Statistics

The estimates of period prevalence (annual prevalence) were expressed as percentage rates with their respective 95% confidence intervals (CI) for the observation year 2020. Insured people who met the case definition formed the numerator for the prevalence estimate. The denominator was made up of all insured people in the study sample. In addition, the prevalence estimates by age, sex and federal state for the population of Germany were standardised according to Statistisches Bundesamt (Destatis; the Federal Statistical Office of Germany) as of 31 December of the respective year (direct standardisation). A control group was formed for the comparison of comorbidities using propensity score matching (1:3 greedy nearest neighbour matching; exact by age, sex and federal state) [29]. The differences between the comorbidities within the populations were visualised using rate ratios (RR) with the respective 95% CI. Baseline characteristics were presented as percentages for categorical variables, while the median and range were used for continuous variables. The analyses were performed with SAS Version 9.4 German (SAS Institute, Cary, NC, USA).

Ethics Statement

The study was conducted according to the principles expressed in the Declaration of Helsinki. We took the criteria of a National Good Practice Guideline into consideration [26, 30]. According to the Good Practice of Secondary Data Analysis, no approval of an ethical committee was required. Permission was granted to access the data used in this study.

Results

Prevalence

In 2020, 49,911 insured persons were documented with at least one acne code (L70). This corresponds to an age- and sex-standardised prevalence rate of about 2.36% (Table 1). Extrapolated to Germany, 1.63 million individuals would be affected. From 2016 to 2020, there was a slight decrease in prevalence of 0.33%. Among these, 1.50% of individuals aged > 25 years were documented with AT (about 950,000 persons in Germany). Following the second definition, 1.31% of individuals aged > 30 years had AT (about 764,000 persons countrywide). AJ was documented in 3.88% of individuals aged ≤ 25 years (about 773,000), or 3.75% of individuals aged ≤ 30 years (about 931,000).

Table 1.

Prevalence of persons with at least one inpatient or outpatient acne diagnosis by observational year and by group [total, acne tarda (a) ≥ 25 and (b) ≥ 30 years and acne juvenilis (a) < 25 and (b) ≤ 30 years]

Year	Age group in years	Insured people (N)	Persons with diagnosis (n)	Rate (%)	95% CI	Standardized rate (%)	Extrapolated to Germany
2016	Total	2,562,928	61,388	2.40	2.38–2.41	2.69	2,175,245
	< 25	522,147	28,470	5.45	5.39–5.51	4.42	875,064
	< 30	660,044	35,904	5.44	5.38–5.49	4.30	1,082,269
	≥ 25	2,040,781	32,918	1.61	1.60–1.63	1.69	1,059,900
	≥ 30	1,902,884	25,484	1.34	1.32–1.36	1.43	820,539
2017	Total	2,459,343	57,662	2.34	2.33–2.36	2.64	1,794,662
	< 25	502,166	26,595	5.30	5.23–5.36	4.31	855,223
	< 30	631,103	33,357	5.29	5.23–5.34	4.18	1,051,655
	≥ 25	1,957,177	31,067	1.59	1.57–1.60	1.66	1,045,852
	≥ 30	1,828,240	24,305	1.33	1.31–1.35	1.42	818,214
2018	Total	2,425,550	54,967	2.27	2.25–2.28	2.55	1,742,179
	< 25	493,680	25,233	5.11	5.05–5.17	4.14	824,407
	< 30	618,110	31,381	5.08	5.02–5.13	4.01	1,005,689
	≥ 25	1,931,870	29,734	1.54	1.52–1.56	1.61	1,017,705
	≥ 30	1,807,440	23,586	1.30	1.29–1.32	1.39	807,277
2019	Total	2,396,254	53,127	2.22	2.20–2.24	2.48	1,711,544
	< 25	487,299	24,283	4.98	4.92–5.04	4.03	804,294
	< 30	607,750	30,053	4.94	4.89–5.00	3.91	977,175
	≥ 25	1,908,955	28,844	1.51	1.49–1.53	1.58	1,001,088
	≥ 30	1,788,504	23,074	1.29	1.27–1.31	1.38	800,829
2020	Total	2,365,230	49,911	2.11	2.09–2.13	2.36	1,630,431
	< 25	482,394	22,996	4.77	4.71–4.83	3.88	772,503
	< 30	598,216	28,189	4.71	4.66–4.77	3.75	930,645
	≥ 25	1,882,836	26,915	1.43	1.41–1.45	1.50	946,258
	≥ 30	1,767,014	21,722	1.23	1.21–1.25	1.31	763,686

CI  confidence interval

The non-specific code L70.9 was used most frequently (52.51%; Fig. S1 in the electronic supplementary material), followed by acne vulgaris (L70.0) in 41.97%.

Prevalence by Sex and Age

In 2020, the average age of insured persons with acne was 31 years [median 26, minimum (min) 0, maximum (max) 100]. Women with documented acne on average were almost 4 years older (mean 32.18 years, median 28, min 0, max 100) than men (mean 28.43 years, median 23, min 0, max 98). The mean age was 18.84 years in AJ (< 30 years) and 46.77 in AT. The sex differences were greater in AT (73.80% in women versus 26.20% in men) than in AJ (64.55% versus 35.45%). At the age of less than 1 year, boys were more frequently diagnosed with (neonatal) acne than girls. People between 10 and 30 years were most likely to have acne with standardised prevalence rates of 6.49% (ages 10–20 years) and of 4.74% (ages 20–30 years). With higher age, the prevalence gradually decreased (Fig. 1).

Fig. 1.

Standardised prevalence of acne (International Statistical Classification of Diseases and Related Health Problems L70) by sex and age groups in 2020

Focusing on people between 0 and 20 years (Fig. 2), the prevalence of acne increased continuously from the age of 10 years to 13.02% by the age of 17 years and then decreased to 8.38% by the age of 20 years. About 6.21% of infants (< 1 year) were diagnosed with neonatal acne.

Fig. 2.

Standardised prevalence of acne (International Statistical Classification of Diseases and Related Health Problems L70) in persons aged between 0 and 20 years in 2020

Prevalence by Federal State

In 2020, the lowest prevalences of AJ were recorded in Bremen, Saarland, Hamburg and Mecklenburg-Western Pomerania, with less than 3% each, and the highest in Brandenburg, Thuringia and Rhineland-Palatinate, with more than 7% (Fig. 3). The regional prevalences of AT were lowest in Bremen, Saxony, Schleswig–Holstein, Saarland and Baden-Württemberg (< 1.3%) and highest in Hamburg, Mecklenburg-Western Pomerania and Saxony-Anhalt (> 1.5%).

Fig. 3.

Standardised prevalence rates of acne juvenilis [left – International Statistical Classification of Diseases and Related Health Problems (ICD) L70 and ≤ 30 years] and of acne tarda (right – ICD L70 and ≥ 30 years) in the claims data set by federal state in 2020. Source: Federal Agency for Cartography and Geodesy: Dataset of administrative areas, shape format 2020 [Available from: https://gdz.bkg.bund.de/index.php/default/verwaltungsgebiete-1-2-500-000-stand-31-12-vg2500-12-31.html (last accessed: 14 January 2021)

Comorbidity

After adjustment by propensity score matching for sex, age and federal state, insured persons with AT were significantly more likely than persons without AT to have the following skin conditions: folliculitis (RR 8.89; CI 7.64–10.35), pyoderma (RR 7.27; CI 5.14–10.28) and rosacea (RR 5.53; CI 5.16–5.93; Fig. 4). Persons with acne were also at higher risk for the following comorbidities not related to the skin: ovarian dysfunction (RR 2.36; CI 2.04–2.73), pervasive developmental disorder (e.g. autism RR 2.20; CI 1.60–3.02) and chronic pancreatitis (RR 2.08; CI 1.75–2.48). Persons with AJ had similar skin-related comorbidities as persons with AT (Fig. S2 in the electronic supplementary material). The sensitivity test for comorbidities in people with acne treated with a specific medication showed only small variations.

Fig. 4.

Top 50 relative risks for predefined comorbidities in people with acne tarda (International Statistical Classification of Diseases and Related Health Problems L70 and ≥ 30 years) compared with persons without acne diagnosis, split by skin disease (top) and non-skin-related comorbidity (bottom)

Drug Prescriptions

In all, 45.29% of people with AJ and 24.57% with AT received at least one acne medication (Table 2). A topical medication was used more often in AJ (87.99%) than in AT (75.10%). By contrast, systemic medication for acne was more often used in AT (34.78%) than in AJ (25.73%). Of the topical medications, anti-infectives were most frequently prescribed in AT (35.19%) and retinoids in AJ (46.06%; Table 2). Between 2017 and 2020, a slight downward trend in prescriptions for topical anti-infectives (except clindamycin and benzoyl peroxide in AT) and an increase in prescriptions for topical retinoids (especially adapalene) was observed (Fig. S3 in the electronic supplementary material).

Table 2.

Prescriptions of acne drugs for persons with AT (≥ 30 years) and AJ (AT N = 21,722; AJ N = 28,189; multiple counting possible) in 2020

Compounds	AT					AJ
	n	Percentage (%)	Mean	Total DDDa	Mean DDDa	n	Percentage (%)	Mean	Total DDDa	Mean DDDa
Total prescriptions	5337	100.00 (24.57)	1.85	269,946	50.58	12,768	100.00 (45.29)	2.11	836,239	65.49
Topical	4008	75.10	1.56	185,126	46.19	11,234	87.99	1.70	656,923	58.48
Retinoids	1455	27.26	1.37	74,362	51.11	5881	46.06	1.50	349,518	59.43
Tretinoin	56	1.05	1.34	1900	33.93	111	0.87	1.22	3700	33.33
Adapalene	287	5.38	1.34	16,645	58.00	1087	8.51	1.37	63,915	58.80
Trifarotene	36	0.67	1.00	2700	75.00	128	1.00	1.05	10,125	79.10
Adapalene and benzoyl peroxide	544	10.19	1.41	30,060	55.26	3237	25.35	1.46	187,695	57.98
Tretinoin, combinations	321	6.01	1.17	15,590	48.57	1242	9.73	1.30	69,317	55.81
Retinol, combinations	296	5.55	1.22	7467	25.23	597	4.68	1.22	14,767	24.74
Peroxides	4	0.07	1.00	113	28.13	60	0.47	1.22	1492	24.87
Anti-infectives	1878	35.19	1.40	78,572	41.84	4820	37.75	1.51	248,581	51.57
Clindamycin	286	5.36	1.34	12,120	42.38	341	2.67	1.38	14,550	42.67
Erythromycin	684	12.82	1.32	14,802	21.64	1210	9.48	1.31	24,828	20.52
Nadifloxacin	190	3.56	1.34	4588	24.14	313	2.45	1.30	6950	22.20
Chlortetracycline (until 2022)	20	0.37	1.05	263	13.13	8	0.06	1.00	113	14.06
Erythromycin, combinations	145	2.72	1.42	9840	67.86	248	1.94	1.40	16,230	65.44
Clindamycin and benzoyl peroxide	624	11.69	1.38	36,960	59.23	2920	22.87	1.52	185,910	63.67
Other	1101	20.63	1.47	32,080	29.14	2143	16.78	1.38	57,332	26.75
Azelaic acid	1101	20.63	1.47	32,080	29.14	2130	16.68	1.38	56,845	26.69
Salicylic acid	–	–	–	–	–	3	0.02	1.33	267	88.89
Bituminosulfonates	–	–	–	–	–	2	0.02	1.00	40	20.00
Bituminosulfonates, combinations	–	–	–	–	–	10	0.08	1.10	180	18.00
Systemic	1856	34.78	1.95	84,820	45.70	3285	25.73	2.37	179,316	54.59
Isotretinoin	460	8.62	3.00	32,097	69.78	1278	10.01	3.67	109,620	85.77
Antibiotics	1437	26.93	1.54	52,314	36.40	2161	16.93	1.43	69,644	32.23
Erythromycin	4	0.07	1.25	65	16.25	6	0.05	1.17	68	11.25
Doxycylclin	901	16.88	1.37	35,006	38.85	1147	8.98	1.28	42,489	37.04
Minocyclin	561	10.51	1.68	16,933	30.18	1052	8.24	1.53	27,040	25.70
Tetracyclin	23	0.43	2.22	720	31.30	7	0.005	1.43	100	14.29

AJ acne juvenilis, AT acne tarda, DDD defined daily dose

Dermatologists were markedly more likely to prescribe medication (AT 65.77% and AJ 76.27%) than other specialist groups (general practitioners: AT 17.29% and AJ 14.31%; paediatricians: AJ 7.4%). Dermatologists most commonly prescribed adapalene in combination with benzoyl peroxide [87.72% in AT and 85.75% in AJ of total defined daily dose (DDD)] and systemic isotretinoin (81.20% in AT and 90.06% in AJ of total DDD), while general practitioners preferred anti-infectives, especially chlortetracycline (52.38% in AT and 44.44% in AJ of total DDD), and systemic antibiotics, especially tetracycline (58.33% in AT and 67.50% in AJ of total DDD; Fig. 5).

Fig. 5.

Volume (defined daily dose in million) of relevant prescribed medications in acne tarda [International Statistical Classification of Diseases and Related Health Problems (ICD) L70 and ≥ 30 years] and acne juvenilis (ICD L70 and ≤ 30 years) by specialist in 2020. Top. topical, Syst. systemic

Discussion

The administrative prevalence of acne in 2020 in the German claims data was 2.36%, which is lower than the previously published global prevalence of 9.4% but in line with a previous German publication of 3.3% and a Danish publication of 3.7% [1, 8, 31]. Other studies have reported a lifetime prevalence between 57.8% and 73.3% [2, 3, 28]. The AT of 1.31% (age ≥ 30 years) was also slightly lower than in previous literature, which ranged between 3% and 4% [7, 32]. Evidently, the comparability of these results is limited because the studies used different methodologies and included different groups of people (e.g. only employees or people aged 12 years and older; 15–18 and 70 years or 35 and 44 years). The lower prevalences compared with the literature may be due to an underestimation in the claims database used in this study. A reason might be that some persons with acne do not seek medical help but rather treat their acne themselves [25, 33]. Furthermore, acne may be misdiagnosed, particularly in adults. Possible differential diagnoses include papulopustular rosacea, folliculitis, syringoma, milia, demodex folliculitis and pityrosporum folliculitis [34]. Conversely, prevalences may have been overestimated in previous primary studies based on patient self-report. However, the study population investigated has been recommended as a good compromise for prevalence estimates compared with cross-sectional population studies and other sources since there is a valid population base [35].

Medicines were not included in the case definition, because some people are not treated with prescription drugs (AJ 45.29% and AT 24.57%), but may use over-the-counter products not represented in the claims data. In addition, some drugs, such as antibiotics and retinoids, are also prescribed for other skin conditions.

The prevalence by age showed that AJ was by far most commonly diagnosed in the 15–20-years age group, particularly among 16- and 17-year-olds. This is also reflected in another previously published study by Wolkenstein et al. (2018), where the highest prevalence also occurred in the 15- and 17-year age group [3]. In terms of sex, the data showed that female participants were much more likely to be affected than male patients, which is also reported in the literature [2, 36]. The sex difference could be due to divergent preferences in the use of cosmetic products and diet, but also other comorbidities such as hormonal diseases and polycystic ovary syndrome, which are more common or specific for women. Cycle-related hormonal fluctuations could also be a reason for the higher burden of acne in women [8].

This work highlights that persons with acne show a large number of comorbidities. As reported in previous literature, individuals with acne are significantly more likely to have some other skin diseases, but also ovarian dysfunction, rhinitis allergica, Crohn’s disease and insomnia, compared with individuals without acne [17–21]. The high proportion of skin-related comorbidities, such as rosacea or folliculitis, may be due to overlapping pathogenesis but could also indicate possible misdiagnosis, which cannot be further investigated with SHI data alone. For this, primary data verified by dermatologists are needed.

With respect to the marked geographical variations, a distinction needs to be drawn between true differences in the prevalence due to still unknown biological and ecological factors and differences due to different treatment or coding habits. To date, both are possible and need further clarification in future studies.

Early and effective treatment, adapted to the severity of acne, can prevent complications such as scarring [37, 38]. In the data presented here, the need for acne treatment is generally underestimated. Anti-infectives, retinoids and oral antibiotics were the most common prescribed drugs for acne, which is in line with a previous study using primary data [25]. Although the treatment of acne has changed over the years (decrease in anti-infectives and increase in retinoids), it is noticeable that some anti-infectives are the most commonly prescribed drugs for AT. It is also clear that guideline-concordant treatments are predominantly prescribed by dermatologists [39]. In particular, the retinoid isotretinoin is rarely prescribed by non-dermatologists, but is strongly recommended for acne not only in advanced cases [40, 41]. This differs from previous recommendations [42, 43] to use isotretinoin only as second-line treatment after unsuccessful topical and oral antibiotic regimens. This change in the evidence-based guideline in the year 2016 for the treatment of acne [39] is not sufficiently communicated to physicians. Moreover, non-dermatologists do often not feel comfortable with the guidance and necessary controls of retinoids, which demand that the patient has good comprehensive information about side effects, e.g., dryness of the lips and eyes and the strict regulations for contraception in women.

Strengths and Limitations

The fundamental strength of this analysis is the large number of SHI data and their scientific added value in terms of population coverage. This large patient cohort provides a comprehensive database without significant selection or recall bias.

However, some limitations must be taken into account when interpreting the results. For example, the population groups of the different health insurance funds differ [44]. To reduce this bias as much as possible, prevalence rates were presented in an age- and sex-adjusted manner. Due to insufficient or inadequate differential diagnoses, misclassifications or coding behaviour of the practitioner, the proportion of patients diagnosed with acne may be either over- or underestimated, which limits the reliability of the statements [45]. External validity testing is needed to assess the extent of insufficient or inappropriate differential diagnosis or coding of historical events (e.g. diseases which no longer exist). For such an assessment, a future linkage of SHI routine data with primary data (e.g. registers) is necessary. This would also enable checking of the reliability and validity of the results obtained.

Claims data do not include important clinical information such as weight, lifestyle and education [26]. Such pre-treatment variables would have been useful for selecting covariates to form accurate propensity scores and for adjusting the data for known confounders [46].

Data on privately paid services are not included, which limits the completeness of the dataset. For example, laser treatments are used for acne, but are not covered by health insurance and are therefore not included [39].

Nevertheless, the health insurance data used, since adjusted and linked with validation criteria, remain an important source for epidemiological and health services research.

Conclusions

Acne is a chronic inflammatory skin condition common in both adolescents and adults, though it is underestimated in SHI data. People with acne have a variety of comorbidities, some of which differ between acne juvenilis (AJ) and acne tarda (AT), and may indicate the need for further pathogenetic investigation and adequate differential diagnosis. The high prevalence of acne in the adult population and its association with a variety of comorbidities is remarkable. Accordingly, acne should be given sufficient attention in routine care beyond adolescence.

In addition, our results show that a third of patients are treated by general practitioners and other non-dermatologists. In a relevant proportion, there is still an undersupply, especially with guideline-based medicines. Interdisciplinary care by specialists such as dermatologists could support adequate treatment and better establish knowledge of guideline in standard acne care. The marked geographical differences in the prevalence of AT/AJ require further clarification in consecutive primary studies. In addition to prescribing behaviour, the duration of medication should also be considered in further analyses to obtain a detailed view of guideline-based care. Due to the highly selected target group and chosen secondary parameters, further multisource-data analysis to connect primary and secondary data will be needed to a receive a complete picture of healthcare and epidemiology of AJ and AT.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Conceptualisation: Kristina Hagenström (lead) and Matthias Augustin (equal). Data curation: Kristina Hagenström (equal) and Katharina Müller (equal). Formal analysis: Kristina Hagenström (equal), Theresa Klinger (equal) and Katharina Müller (equal). Funding acquisition: Matthias Augustin (lead). Investigation: Katharina Müller (equal) and Matthias Augustin (equal). Methodology: Kristina Hagenström (lead), Theresa Klinger (equal) and Matthias Augustin (equal). Project administration: Theresa Klinger (equal). Resources: Kristina Hagenström (equal). Software: Katharina Müller (equal). Supervision: Kristina Hagenström (lead). Validation: Kristina Hagenström (lead) and Katharina Müller (equal). Visualisation: Kristina Hagenström (lead) and Katharina Müller (equal). Writing – original draft preparation: Kristina Hagenström (lead), Theresa Klinger (equal) and Matthias Augustin (equal). Writing – review and editing: Kristina Hagenström (equal), Theresa Klinger (supporting); Brigitte Stephan (supporting) and Matthias Augustin (equal).

Funding

The project was financially supported by Galderma Laboratorium GmbH, Düsseldorf, Deutschland. Furthermore, we also acknowledge the financial support for the publication costs from the Open Access Publication Fund of Universitätsklinikum Hamburg-Eppendorf (UKE) and the German Research Foundation (DFG).

Data Availability

The datasets generated for the claims data cohort are not available, as the use of claims data is restricted to defined persons.

Declarations

Conflict of Interest

Kristina Hagenström, Katharina Müller and Theresa Klinger declare no conflicts of interest. Brigitte Stephan has received payments/honoraria for lectures and presentations and/or received grants and/or participated in clinical trials including from the following companies: AbbVie, Almirall Hermal, Amgen, Beiersdorf, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Glaxo SmithKline, Janssen-Cilag, LEO Pharma, Lilly, Medac, Novartis, Pierre Fabre, Sanofi Aventis and UCB. Matthias Augustin has received institutional research funding, travel grants and lecture fees from companies which produce products for acne, including Almirall, Beiersdorf, Galderma and GSKx.

Ethical Approval

The study was conducted according to the principles expressed in the Declaration of Helsinki. We took the criteria of a National Good Practice Guideline into consideration [26, 30]. According to the Good Practice of Secondary Data Analysis, no approval of an ethical committee is required. Permission was granted to access the data used in this study.