Self-Report of Chronic Diseases

Self-Report of Chronic Diseases

LASA filenames:
LASA035 / LASA235 / LASA435
LASA602 / LASA702
LASEt602 / LASEt702

Contact: Natasja van Schoor


In the conceptual model that forms the basis of the LASA-design, chronic diseases are included as independent variables or, in other words, as predictors of changes in different aspects of functioning of older people. Because chronic diseases are not considered as one of the main outcomes in the model, it was considered adequate to conduct the assessment of the presence of chronic diseases mainly by respondents’ self-reports.

Questions concerning the presence, treatment and severity of selected chronic diseases are included in the main interview (LASA*035). The selection of chronic diseases that were explicitly investigated was based on prevalence (the most frequently occurring somatic chronic diseases in the Netherlands; roughly >5.0% in the age group 55+ [Van den Hoogen et al. 1985]) and functional consequences:
1) chronic non-specific lung disease (=CNSLD= obstructive lung disease (OLD)) = asthma or chronic obstructive pulmonary disease (=COPD)), 2) cardiac disease, 3) peripheral arterial disease (PAD), 4) diabetes mellitus (DM), 5) cerebrovascular accident (CVA) or stroke, 6) osteoarthritis (OA), 7) rheumatoid arthritis (RA), and 8) cancer. In addition, respondents were asked whether they had any other chronic diseases (defined as a disease of which symptoms and/or treatment had been present for at least three months). A maximum of two other chronic diseases could be recorded during the interview. Apart from this, it was explicitly asked whether the respondent had involuntary urine-loss, although this is not considered a chronic disease but, instead, an impairment. Furthermore, from the C-cycle (LASAC035) onwards hypertension was explicitly asked as a chronic disease, and from the D-cycle (LASAD035) onwards head trauma was explicitly asked as well (for the documentation on Head trauma, see here).

Presence of chronic diseases

The ‘core’ questions concerning the presence or absence of the eight specific chronic diseases were derived from the chronic disease questionnaire of Statistics Netherlands [CBS 1989] and selected based on discussions with experts. These core questions were also included in the short version of the main interview. From the C-cycle they are included in the telephone survey among respondents or proxy’s. For the C-cycle, they can be found in file LASAC602, and file LASAC600 or LASAC002 can be used to distinguish between who was interviewed: the respondent or proxy. From the D-cycle the telephone survey was cut into two different telephone surveys: one with the respondent (LASAD702) and one with the proxy (LASAD602).

In 1992/1993, 2001/2002, 2005/2006, 2010 and 2016, information on chronic diseases was obtained from General Practitioners (GPs) for the period since the last GP approach. Information on the GP data collections can be found in the documentation files found under ‘GP information

In comparison with GP information, the self-reports of the chronic diseases asked explicitly were found to be fairly accurate, with the exception of arthritis and PAD. Osteoarthritis and rheumatoid arthritis were combined because it appeared to be difficult for respondents to distinguish these two diseases, especially for those with osteoarthritis. Kappa’s ranged from 0.30 to 0.40 for arthritis and PAD, to 0.85 for DM [Kriegsman et al. 1996]. This accuracy was assessed for respondents who completed the full interview at baseline (B-cycle). For the G-cycle (2008-2009), the accuracy was assessed again, and found to be almost identical to that in 1992-1993, although for cardiac disease, arthritis and cancer more overreporting was observed [Galenkamp et al 2014].

In the main interview, when a respondent reported a chronic disease to be present, branching questions concerning the specific disease were asked (‘disease specific questions’). Standard additional questions concerned medical treatment (use of medication and regular contact with a physician) and disease-specific symptoms and signs. For each specific disease, pertinent medical history questions (questions that physicians use to get an impression of disease-severity) were selected in close cooperation with general practitioners with expert knowledge on the specific chronic diseases. For malignancies, an oncologist was consulted [Kriegsman et al 1997].

Response categories

At the B-cycle the response categories were: Missing, No and Yes. At the C-cycle and later cycles, the No and Yes response categories of the presence of a chronic disease were extended as follows: ‘No, never’; ‘No, Yes at previous cycle’; ‘Yes, No at previous cycle’ and ‘Yes, Yes at previous cycle’. Until the D-cycle, it was recommended to recode the response categories as: Missing, ‘No, never’ as No, and the other three categories as Yes. It was assumed that once a respondent has a chronic disease, (s)he is affected for the rest of his or her life. From the E-cycle onwards, the chronic diseases were not all seen as chronic due to the major medical developments of the last few years. For example; patients with plaque in the coronary arteries can be cured by angioplasty (or dottering) and patients with breast cancer can be cured by radiation, chemotherapy and/or surgery. In the arthritis algorithm it was chosen to recode the two response categories starting with ‘No,..’ as No, and the two categories starting with ‘Yes,..’ as Yes.

Use of information on chronic diseases


It is recommended to combine the information on chronic diseases from respondents not participating in the main interview, but participating in the telephone surveys (LASA*602, LASA*702) with the information included in the files of the main interviews before starting analyses.

Summary measure: number of chronic diseases

The number of chronic diseases (data from main interview: LASA235) is primarily used as an independent variable, measuring the presence of (co)morbidity. In constructing this variable, osteoarthritis and rheumatoid arthritis are considered as one disease.

Depending on the specific purpose, two ways of counting the number of chronic diseases can be used:
– *nochrom: the number of chronic diseases that were explicitly asked about (including CNSLD, cardiac disease, PAD, stroke, DM, arthritis and malignancies), ranging from 0-7.
– *nochrot: the number of chronic diseases that were explicitly asked about, including the seven diseases above, hypertension and other diseases, but not incontinence and head trauma, ranging from 0-9. Note: respondents may report non-chronic diseases at the “other disease” question, e.g. TIA. These diseases are also included in the *nochrot variable. In addition, at wave K we included an extra question on hypercholesterolemia, which was not included in the *nochrot variable. However, it is possible that hypercholesterolemia was reported by some respondents at the “other disease” question before wave K and thus included in *nochrot before wave K.

Use of branching questions
Additional questions are asked when a respondent answers that a specific disease is present. These branching questions can be used in different ways, serving different purposes. First, because of the possibility that self-reports tend to overreporting [Galenkamp et al 2014], the standard branching questions can be applied by the researcher to more narrowly define the presence of a chronic disease [Portrait et al 1999]. Thus, a chronic disease may be considered to be present only when under continuous medical treatment, i.e., when the patient reports to use prescribed drugs, and/or to have regular contact with a physician (general practitioner or medical specialist) for this disease. This narrower definition, however, was shown to have hardly any influence on the accuracy of respondents’ self-reports [Kriegsman et al. 1996]. Second, the branching questions can be used to improve the prediction of outcome measures such as physical functioning [Kriegsman et al. 1997]. Third, the branching questions can be used to define subgroups within a specific chronic disease, for example, according to disease severity [Penninx et al. 1997].

Symptoms of CNSLD include the presence of daily coughing, daily phlegm production, periods of increased coughing or phlegm-production during the past year, shortness of breath with light exertion or at rest, and wheezing breath during rest, and the degree of disturbance of night-rest because of CNSLD symptoms. Disease-specific questions were partly derived from the questionnaire developed by Van der Lende et al. [Lende 1975].

For cardiac diseases, the responses to the disease-specific questions, which were partly adapted from the Rose-questionnaire [Rose 1962], can be used to distinguish separate diagnoses. A history of myocardial infarction (MI) can be considered present when the patient positively responded to the question whether he or she ever had a MI. Angina pectoris (AP) can be considered present when the patient reported to experience pain, or a heavy, uncomfortable feeling on the chest during exertion (walking stairs or walking or cycling against the wind), disappearing within 10 minutes after stopping or taking sublingual nitroglycerine. Also, when the patient avoids exertion because of chest pain, AP can be considered present. Other coronary artery disease can be defined as a history of coronary artery surgery (Coronary arterial bypass grafting (CABG)), without MI and without symptoms of AP. Congestive heart failure can be considered present when the patient has to sleep with more than one pillow because of shortness of breath and/or reports to have edema of the ankles, feet or legs at the end of the day (the latter symptom is asked for the first time at the J-cycle). Diseases of the cardiac valves and serious arrhythmia’s are considered present when the patient reports to have had valvular surgery or a pacemaker implantation, respectively. All patients, who do not fit in at least one of the diagnostic categories described above, can be considered to have ‘other cardiac disease’. This category thus includes, for instance, relatively benign arrhythmias or valvular lesions without a history of surgery. Additionally, in a subsample at LASA-C and LASA-D, the degree of aortic calcification was assessed with lateral radiographs.

For PAD, the following symptoms are distinguished: intermittent claudication (experience of pain or cramps in the calves when walking which disappears within 10 minutes after standing still), history of arterial surgery (bypass or dotter procedure) of the abdomen or lower limbs, and a history of other arterial surgery.

The symptoms of DM pertain to the presence of diabetic complications: macrovascular complications (AP and/or intermittent claudication), retinopathy, and peripheral neuropathy. Macrovascular complications are defined as the presence of symptoms indicating AP and/or intermittent claudication. These are considered complications only when the patient does not report cardiac disease and PAD, respectively. Retinopathy is considered present if the patient reports a history of laser coagulation therapy. Peripheral neuropathy can be considered present when pain in the legs and/or feet is reported during rest.

In cardiac diseases, PAD and DM, some disease-specific questions are the same and were partly adapted from the Rose-questionnaire [Rose 1962]. If a person reported both DM and cardiac disease or PAD, the questions which were the same were asked only once.

For stroke, the presence of specific sequelae is included, namely locomotor disabilities, visual handicap, and expressive (such as difficulties in finding words) or perceptive (understanding written text) aphasia. A history of two or more strokes is included separately.

For arthritis (osteoarthritis and/or rheumatoid arthritis), both the presence of symptoms and the extent of joint involvement, as well as a history of joint surgery are taken into account. The following symptoms are included: pain in one or more joints during most days of the previous three months, morning-stiffness in the joints during most days of the previous three months, and swelling of one or more joints during most days of the previous month. Upper body involvement is present when symptoms are experienced in the hands, wrists, elbows, shoulders or neck. Lower body involvement includes the presence of symptoms in toes, feet, ankles, knees or hips. Because the symptoms and the functional consequences of arthritis, especially when confined to few joints, can be alleviated by joint replacement surgery, the presence of a history of surgery of the joints of the upper, respectively the lower body, is asked separately.
Moreover, in a subsample at LASA-C and LASA-D, spinal disc degeneration was assessed with lateral radiographs by using the four point Kellgren scale (Kellgren et al., 1963) (see documentation on Lateral radiographs: vertebral deformities, disc degeneration, and aortic calcification; LASAC805/LASAD805 by S. Pluijm & N. van Schoor).

For cancer age of diagnosis, localization and receiving or having received treatment for cancer are taken into account. Cancer localization is categorized as lungs, breast, uterus, cervix, intestines, prostate, larynx, eosophagus, stomach, blood, skin or other. Metastases and the localization of the metastases (bone, liver, brain, lungs, lymph nodes, other) are also covered.

Disease algorithms

For several diseases, it was considered desirable to have a more objective diagnosis and more specificity. To this end, the branching questions can be combined with information from other data sources (medical interview: inspection of medicine bottles, and medical records of general practitioners) in LASA. Penninx et al. combined these sources to obtain a diagnosis for cardiac disease [Penninx et al. 2001]. Moreover, algorithms (decision trees) were developed for highly prevalent cardiac diseases: AP, MI, Congestive Heart Failure (CHF) and Cardiac Arrhythmia (AR) and for major vascular diseases: PAD and CVA (or stroke) [Bremmer et al. 2006, Schalk et al. 2004]. Algorithms were developed for diabetes mellitus as well. Finally, algorithms were developed for the prevalence and incidence of osteoarthritis, and for rheumatic disease. In 2018, the algorithms for cardiac disease, major vascular diseases and diabetes mellitus were revised because of new insights.

Longitudinal data on chronic diseases (B- and C-cycle only)

Based on additional information from the respondent in consecutive measurement cycles, the information from previous cycles may have been adapted during longitudinal cleaning. For example: when the respondent indicated chest complaints who were investigated by a cardiologist, but no cardiac disease has been diagnosed, the respondent had thus for the B-cycle on bhart01= Yes. If during the C-cycle the respondent indicates that three years ago his chest complaints were caused by heartburn, the bhart01 on the B-cycle will be changed into a No. The longitudinally cleaned data on chronic diseases for the B and C measurement cycles are available in the files LASAB435 and LASAC435. All longitudinally cleaned variables start with a x (for example xcara01, xnochrot). Be aware that data of the D-cycle and later cycles are not cleaned longitudinally. For cross-sectional analyses the files LASA*035 and LASA*235 should be used.

Keep in mind that the data are based on self-report, and assessed as predictors, not as outcomes. However, the incidence of a chronic disease can be determined, when combined with other data sources (medical interview: inspection of medicine bottles, and medical records of general practitioners), by assessing whether respondents not reporting the disease at a particular measurement cycle, report this disease at the consecutive cycle. In assessing the incidence of a chronic disease, longitudinal decisions should be made per disease.

LASAB035 / LASAC035 / LASAD035 / LASAE035 / LAS2B035 / LASAF035 / LASAG035 / LASAH035 / LAS3B035 / LASMB035 / LASAI035 / LASAJ035 / LASAK035 (main interview, in Dutch);
LASAC602 / LASAD602 / LASAE602 / LASAF602 / LASAG602 / LASAH602, LASAI602 / LASAJ602 / LASAK602 (telephone interview with PROXY, in Dutch);
LASAC702 / LASAD702 / LASAE702 / LASAF702 / LASAG702 / LASAH702, LASAI702 / LASAJ702 / LASAK702 (telephone interview with RESP, in Dutch)

Interim measurement:

LASEs802 (self-admin. questionnaire, in Dutch);
LASEt602 (telephone interview with PROXY, in Dutch);
LASEt702 (telephone interview with RESP, in Dutch)

Variable information
LASAB035 / LASAC035 / LASAD035 / LASAE035 / LAS2B035 / LASAF035 / LASAG035 / LASAH035 / LAS3B035 / LASMB035 / LASAI035 / LASAJ035 / LASAK035 (K not available yet);
LASAB235 / LASAC235 / LASAD235 / LASAE235 / LAS2B235 / LASAF235 / LASAG235 / LASAH235 / LAS3B235 / LASMB235 / LASAI235 / LASAJ235 / LASAK235 (K not available yet) (nr. of diseases);
LASAB435 / LASAC435 (longitudinally cleaned)
LASAC602 / LASAD602 / LASAE602 / LASAF602 / LASAG602 / LASAH602 / LASAI602 / LASAJ602 / LASAK602 (K not available yet)
LASAC702 / LASAD702 / LASAE702 / LASAF702 / LASAG702 / LASAH702, LASAI702 / LASAJ702 / LASAK702 (K not available yet)

Interim measurement:

(pdf, upon request)

¥ head trauma data (questionnaire, variable information) are published elsewhere, see documentation on Head trauma.

Availability of information per wave

Chronic diseases (head trauma:
see documentation on Head trauma)

Chronic diseases (self-report of -) → 035,
- number of → 235,
- longitudinally cleaned → 435 (only in B and C)
Chronic diseases (self-report of -) → s802----Sa---------
Chronic diseases [PROXY] → 602-TpTpTp--TpTpTp--TpTpTp
Chronic diseases [PROXY] → t602----Tp---------
Chronic diseases [RESP] → 702-TrTrTr--TrTrTr--TrTrTr
Chronic diseases [RESP] → t702----Tr---------

More information about the LASA data collection waves is available here.

* IM=interim measurement between E and F ( first cohort only)
2B=baseline second cohort;
3B=baseline third cohort;
MB=migrants: baseline first cohort;
K=not available yet

Ma=data collected in main interview;
Sa=data collected in self-admin. questionnaire;
Tr=data collected in telephone interview with respondent;
Tp=data collected in telephone interview with proxy

Previous use in LASA

Chronic diseases as main determinant of an outcome

As was envisaged in the LASA-design, numerous articles have used chronic morbidity as a determinant, whether as separate diseases, as the number of diseases, or as multimorbidity (the presence of two or more diseases). For example, in prospective study designs, Kriegsman et al [2004] examined how combinations of diseases affected functional decline, Comijs et al [2009] examined how single diseases affected decline in cognitive functioning, Bisschop et al [2004] examined the longitudinal association of single diseases and depression, and Van Gool et al [2007] examined lifestyle transitions in specific disease groups. In a study in a subsample of LASA, it was shown that a history of cardiovascular disease (CVD) raised the stress level as derived from hair cortisol [Manenschijn et al 2013]. In a study among older workers, the chance of stopping with paid work was higher for workers with chronic diseases if they had a high physical work load or lacked psychosocial support [Boot et al 2014].

Specific diseases and their correlates have been studied in particular osteo-arthritis in the EPOSA project [e.g., Zambon et al 2016, see also], and Chronic Obstructive Pulmonary Disease [Smid et al 2017]

Chronic diseases were also included in the wider concept of frailty, i.e., the Frailty Index [Hoogendijk et al 2017]. Frailty is considered to be a predictor of unfavourable outcomes (see documentation on Frailty).

Chronic diseases as outcomes

When used as an outcome, several studies have addressed predictors of CVD. In studies of incident CVD, depression [Bremmer et al 2006], low serum albumin [Schalk et al 2006], and the combination of the metabolic syndrome and high IGF-1 levels [Van Bunderen et al 2013] showed an increased risk of CVD. In several studies from the Emerging Risk Factors Collaboration, incidence and mortality of CVD were associated with CRP, metabolic risk factors, blood lipids, and adiposity measures [e.g., Singh et al 2013].

Trends in chronic diseases in relation to other outcomes over time

Chronic morbidity showed a steady increase from 1992 to 2009 [Galenkamp et al 2013]. However, self-rated health showed a stable trend. This discrepancy was explained by the lesser importance attached to chronic morbidity in the perception of one’s health. In contrast, chronic diseases were not observed to become less disabling during this period [Hoeymans et al 2012]. The increase in chronic morbidity, and in particular associated polypharmacy, partly explained the observed increase in acute overnight hospital admissions between 1997 and 2007 [Galenkamp et al 2016]. The trend in mortality of a specific disease, coronary heart disease (CHD), was examined in a collaborative study [Koopman et al 2016]. The decline in CHD-mortality was explained by equally large contributions from improvements in treatments and improvements in population risk factors.


  1. Central Bureau of Statistics (CBS) (1989). Health Interview Questionnaire. Heerlen, the Netherlands: Central Bureau of Statistics.
  2. Hoogen HJM van den, Huygen FJA, Schellekens JWG, Straat JM, Velden HGM van de (1985). Morbidity figures from general practice, data from four general practices 1978-1982. Ned Tijdschr Geneeskd. 1986;130:2101.
  3. Kriegsman DMW, Penninx BWJH, Eijk JThM van, Boeke AJP, Deeg DJH (1996). Self-reports and general practitioner information on the presence of chronic diseases in community-dwelling elderly: A study on the accuracy of patients’ self-reports and on determinants of inaccuracy. J ClinEpidemiol, 49:1407-17.
  4. Galenkamp H, Huisman M, Braam AW, Schellevis FG, Deeg DJH (2014). Disease prevalence based on older people’s self-reports increased, but patient-general practitioner agreement remained stable, 1992-2009. J Clin Epidemiol, 67(7): 773-780. doi: 10.1016/j.jclinepi.2014.02.002.
  5. Kriegsman DMW, Deeg DJH, Eijk JThM van, Penninx BWJH, Boeke AJP (1997). Do disease specific characteristics add to the explanation of mobility limitations in patients with different chronic diseases? A study in The Netherlands. J Epidemiol Community Health, 51:676-85.
  6. Portrait F, Lindeboom M, Deeg DJH (1999). Health and mortality of the elderly: the grade of membership method, classification and determination. Health Econ, 8: 441–457.
  7. Lende R van der, Jansen-Koster EJ, Knijpstra S, Meinesz AF, Wever AMJ, Orie NGM (1975). Definitie van CARA in epidemiologie en preventie. [Definition of CNSLD in epidemiology and prevention]. Ned Tijdschr Geneeskd, 119:1975-1987.
  8. Kellgren JH, Jeffrey MR, Ball J. The Epidemiology of Chronic Rheumatism: Atlas of Standard Radiographs. 2nd Edition. Blackwell Scientific, Oxford, 1963.
  9. Penninx BWJH, Tilburg TG van, Deeg DJH, Kriegsman DMW, Boeke AJP, Eijk JThM van (1997). Direct and buffer effects of social support and personal coping resources in individuals with arthritis. SocSci Med, 44:393-402.
  10. Penninx BWJH, Beekman ATF, Honig A, Deeg DJH, Schoevers RA, Eijk JThM van, Tilburg W van (2001). Depression and cardiac mortality: results from a community-based longitudinal study. Arch Gen Psychiatry, 58:221-7.
  11. Perez CA, Brady LW (eds) (1992). Principles and practice of radiation oncology. Philadelphia: Lippincott.
  12. Rose GA (1962). The diagnosis of ischaemic heart pain and intermittant claudication in field surveys. Bull World Health Org, 27:645-658.
  13. Schalk BW, Visser M, Bremmer MA, Penninx BW, Bouter LM, Deeg DJ (2006). Change of serum albumin and risk of cardiovascular disease and all-cause mortality: Longitudinal Aging Study Amsterdam. Am J Epidemiol. 164(10):969-77.
  14. Bremmer MA, Hoogendijk WJ, Deeg DJ, Schoevers RA, Schalk BW, Beekman AT (2006). Depression in older age is a risk factor for first ischemic cardiac events. Am J Geriatr Psychiatry. 14(6): 523-30.
  15. Galenkamp H, Braam AW, Huisman M, Deeg DJ (2013). Seventeen-year time trend in poor self-rated health in older adults: changing contributions of chronic diseases and disability. Eur J Public Health. 23(3): 511-517. DOI: 10.1093/eurpub/cks031.
  16. Hoeymans N, Wong A, van Gool CH, Deeg DJ, Nusselder WJ, de Klerk MM, van Boxtel MP, Picavet HS (2012). The Disabling Effect of Diseases: A Study on Trends in Diseases, Activity Limitations, and Their Interrelationships. Am J Public Health. 102(1): 163-170.

Date of last update: April 20, 2020