Table B11Evidence Table Of Studies Examining Adverse Effects of Screening (KQ4)

Study
Author, year
Quality rating
Study designPurpose of studyCountry;
Setting;
Year(s) of study
Treatment groups;
Sample size
Length of follow-upInclusion criteriaExclusion criteriaParticipant selectionPopulationSES or educational levelPre-existing depression, anxiety analyzed, etcExisting vascular diseaseFBG (mg/dl)
A1c (%)
Lipids (mg/dl)Blood pressure (mm Hg)Other risk factors (CVD, etc)Measures usedInterventionPrimary endpoint(s)Outcomes for standardized measuresAdherence
withdrawals (%)
Other resultsCommentsFunding
ADDITION Study
Thoolen et al, 2006188
Not rated
2X2 factorial (based on time since diagnosis and treatment intensity) cross-sectional studyInvestigate how time since diagnosis and treatment intensity influences psychological outcomes in patients with screen-detected DM2Southwest Netherlands Multi-center (79 general practices)468 invited
227 agreed
206 completed questionnaire
196 included in analysis (10 not included because time since diagnosis occurred between 1–2y, so did not fit parameters)
No follow-upPatients included in Dutch arm of ADDITION study without serious comorbidities
Ages 50–69
Diagnosed with DM2 3–33m previously
Receiving usual or intensive treatment

From ADDITION STUDY:
Screening study:
Without known DM2
Identified though specific centers
Treatment study: Newly diagnosed DM2, defined by 99 mg/dl (5.5 mmol/l), by fasting and 2-h post-glucose-challenge blood glucose measurements
From ADDITION STUDY:
Screening study:
Previously diagnosed DM2
Treated with blood glucose lowering agents
Treatment study:
IGT and IFG, contraindications or intolerance to study medications, alcoholism, drug abuse, psychosis or emotional problems, malignant disease with a poor prognosis, pregnant or lactating
Screen-detectedMean age: 61–62y
(~5y SD)

% male (“marginal difference” between groups, ns):
Group 1: 71
Group 2: 50
Group 3: 63
Group 4: 57
Educational level*:
Group 1: 3.0±1.6
Group 2: 3.0±1.4
Group 3: 3.4±1.6
Group 4: 3.0±1.7

*Measured on a 6 point scale (1=primary to 6=higher education)
NRNRNRNRNRBMI (from self-report) mean (SD):
Group 1: 29.0 (4.3)
Group 2: 29.4 (4.7)
Group 3: 30.0 (4.9)
Group 4: 30.0 (4.9)
Hospital Anxiety and Depression Scale (HADS): measure emotional outcomes, including anxiety and depression [standardized]

Problem Areas in Diabetes (PAID) Scale: measure diabetes distress [standardized]

Cognitive variables included:
1)

perceptions of health threat - measured by a) perceived seriousness of [based on Diabetes Illness Representations Questionnaire], and b) vulnerability for diabetes [not standardized]

2)

self-efficacy - measured by combination of a) Lorig 1996, and b) Kuijer and de Ridder 2003 scales [not standardized]



Self-care behavior measured using revised summary of diabetes self-care activities measure [parts valid]
4 groups created by categories of usual or intensive multifactorial drug treatment and time since diagnosis (<1 y or 2–3 y)

Multivariate analysis used to examine variation in outcomes on time since diagnosis and treatment intensity

4 groups analyzed:
Group 1: DM <1y time since diagnosis + usual care
Group 2: DM <1y time since diagnosis + intensive treatment
Group 3: DM 2–3y time since diagnosis + usual care
Group 4: DM 2–3y time since diagnosis + intensive treatment
7 variables: Anxiety, depression, diabetes-related distress, perceived seriousness and vulnerability, self efficacy, and self-care“Most patients reported little distress, low perceived seriousness and vulnerability, high self-efficacy, and low self-care, but outcomes varied considerably across conditions”

Time effects found for perceived vulnerability (increases significantly with time since diagnosis) (F=14.3, p<0.001)

No time effects found for anxiety (F=0.3, ns) nor depression (F=1.2, ns)

No time effects found for diabetes distress (F=3.0, ns), perceived seriousness (F=1.8, ns), self efficacy (F=0.2, ns), nor self management (F=0.0, ns)

Some reported clinically relevant anxiety (HADS score >8; clinically definite scores >11) in group diagnosed < 1 year, but it seems to be effect of intensive treatment × time, because the intensive treatment group is significantly higher (mean scores, 6.8 vs 4.5, F=5.8, p<0.001). 2–3y group mean scores = 5.0 vs 5.5, ns
NRRelated to treatment:

Time × treatment interactions found for anxiety (F=5.8, p<0.01), diabetes-related distress (F=4.6, p<0.05), and self-efficacy (intensively treated patients showed more distress and less self-efficacy in 1st y; usual care patients reported more distress and less self-efficacy 2–3y after diagnosis (F=7.1, p<0.01)
Included participants were more educated and reported lower self-management than non-participants

Analysis adjusted for sex, BMI, and number of complaints

Psychological effects were not associated with sociodemographic variables, but were associated with BMI and medical complaints
NR
ADDITION Study
Eborall et al, 2007190
Fair
Controlled clinical trial (embedded within the ADDITION Trial)To quantify the psychological impact of primary care-based stepwise screening for DM2United Kingdom (Cambridge)Screened: 4370
Control: 964
Up at 15mFrom ADDITION screening study:
Without known DM2
Identified though specific clinical centers
See aboveRecruitment from clinical settings65% male
Mean age: 58y
Avg BMI: 30.5
NSD between groups
NR for these specific groups (see above)NRNRNRNRNRBMI >30kg/m2: (mean [SD])
I: 30.5 (4.7)
C:30.6 (4.9)
Spielberger state anxiety inventory, range from 20–80
Hospital Anxiety and Depression Scale (HADS): measure emotional outcomes, including anxiety and depression [standardized]
Single item on general health
Disease-specific worry: adapted from legman cancer worry scale: sum scores 6–24
Step-wise screening for DM2: hi-risk for DM2 were identified using computerized general practice records; those were invited to get random BG; if >5.5 mmol/l invited for fasting BG, if >6.1 mmol/l invited for 75-g OGTTState anxiety, anxiety, depression, diabetes-specific worry, self-rated healthConclusion: screening has limited psychological impact on patients; being required to return for further tests after an initial positive random BG has small negative psychological impact of doubtful clinical significance

Immediate impact of initial (+) screening test compared to test (-): poorer health; higher anxiety, depression, diabetes-specific worry (p all ≤ 0.05)
Invited to screening and did not attend; 32%
Random BG (-) at baseline: 67% follow-up at 12–15m
Random BG (+) at baseline: 39% follow-up at 12–15m
Test for trend over steps in screening process: worry about DM increased as underwent more screening tests before testing (-)

Nonattenders for the initial test: 11% response rate at 12–15m: had high worry at 12–15m (p=0.03)
Wellcome trust, National Health Service Research and Development
ADDITION Study
Eborall et al, 2007189
Not rated
Prospective qualitative interview of patients in a screening program for DM2To provide insight into factors that contribute to the anxiety reported in the quantitative study of the psychological effect of screening for DM2; to explore expectations and reactions to the screening experienceUnited Kingdom (Cambridge)23 totalNo follow-upFrom ADDITION screening study:
Without known DM2
Identified though specific clinical centers
See aboveRecruitment from clinical settingsPopulation scheduled for OGTT was sampled; additional sampling to address imbalance of sex and diagnosis with initial samplingNRNRNRNRNRNRNROpen-ended questionsAs abovePerceptions and expectations before and after OGTTInitial stages of screening processes: most participants not very worried who tested (+) on the first tests
Prediagnostic test expectations: many accepted possibility of (+) diagnosis
Reactions after new diagnosis of DM2: tendency to downplay importance; all had plans to control the disease; most were grateful for screening program
Diagnosed with IFG or IGT: many were confused by this diagnosis; most were unconcerned and unaware of this diagnosis as a risk factor for DM2 or CVD
NoneNoneRoyal College of General Practitioners scientific foundation board for this study; ADDITION funded by Wellcome trust, National Health Service Research and Development
Edelman et al, 2002182
Good
Cohort with comparison (nondiabetic) groupDetermine effects of new diagnosis of DM2 discovered by systematic screeningUnited States
Durham Veterans Affairs Medical Center, North Carolina (single center)
October 1996 – March 1999
1253 total
(1,177 without DM2 at screening; 56 [4.5%] with new diagnosis of DM2 at screening)
1yDurham Veterans Affairs Medical Center outpatients that did not report having diabetes at start of studyKnown diabetes Patients who had a prescription filled for hypoglycemic medication
Short life expectancy (incurable cancer, heart or lung disease requiring oxygen)
No easy access to a telephone
Systematically screened for DM2Ages: 55y mean (6y SD)
94% male
Race:
69% Caucasian
29% African American
2% Other
NRYesNRNRNRNRBody weight: 60% > 120% of ideal body weight

Comorbidity: 95% comorbid illness; 34% moderate to severe comorbidity
Prior to study, A1c measurements taken on all subjects: A1c > 6.0% were repeated

DM2 defined as A1c > 7.0% or fasting plasma glucose > 126 mg/dl (7.0 mmol/l)

Health-related quality of life (HRQoL) assessed using Medical Outcomes Study Short Form 36 (SF-36). 2 parts: Physical Component Scale (PCS) and Mental Component Scale (MCS)

Comorbidity assessed using Kaplan-Feinstein Index
HRQoL measured at baseline and 1y after diagnosis using multivariate analysisHRQoLNo significant differences (p<0.05) between patients with and without DM2 nor between baseline and 1y follow-up

Baseline PCS:
NonDM2 vs. newly diagnosed DM2 (36.3 vs. 35.6) not different (p=0.67) Baseline MCS:
NonDM2 vs. newly diagnosed DM2 (49.6 vs. 48.8) not different (p=0.70)
1y follow-up PCS:
NonDM2 vs. newly diagnosed DM2 (35.2 vs. 34.6) not different (p=0.68)
1y follow-up MCS:
NonDM2 vs. newly diagnosed DM2 (48.2 vs. 48.0) not different (p=0.94)
NRMild-severe comorbid illness associated with lower PCS both at baseline and 1y follow-up (p<0.05)Supported by Department of Veteran's Affairs Cooperative Studies and a Research Career Award
Farmer et al, 2003183
Good-fair
Single-group cohortTo assess changes in anxiety, well-being, and cognitions associated with screening for DM2 in people at increased risk of DM2 after 1y to identify potential predictors of increased anxiety and lower well-beingUnited Kingdom, Oxfordshire and South Northamptonshire
1996 - 1998
431 total1yProbands:
Age ≥ 35 at diagnosis
Families with ≥ 3 siblings, and a quarter of families with 2 siblings living within study area
Participants:
Participants aged 35–74
Family history of DM2
Not known to have DM2
Able to complete questionnaires
Participants:
Known DM2
< age 35 or > age 74
Recruited with information from general practitioners
Probands sent questionnaires to assess willingness of siblings to participate
Mean age (SD) & % male:
Normal risk of DM: 57.3y (10.2y) & 38.8%
Borderline risk of DM: 59.8y (8.9y) & 48.5%
High risk of DM: 59.8y (9.0y) & 56.5%
Possible diabetes: 58.7y (9.0y) & 72.2%
Occupational group (manual/professional %):
Normal risk of DM: 139/86
Borderline risk of DM: 61/38
High risk of DM: 50/31
Possible diabetes: 12/6
YesNRNRNRNRMean BMI (SD):
Normal risk of DM: 27.3 (5.3)
Borderline risk of DM: 28.4 (4.6)
High risk of DM: 29.9 (5.3)
Possible diabetes: 31.6 (5.9)
Response rates calculated
Speilberger State Anxiety Inventory (SSAI-SF)
Well-being questionnaire (WBQ-12)
Health Anxiety Inventory (HAI)
Questionnaires at baseline and 1y follow-up

Analysis separated according to those receiving a “normal” test result <5.5 mmol/L compared with those “at risk” receiving a borderline (99–108 mg/dl [5.5–6.0 mmol/L]), high (>108–140 mg/dl [>6.0–7.8 mmol/L]), or test result indicating diabetes (140 mg/dl [>7.8 mmol/L])
Anxiety
Well-being
Cognition
Anxiety decreased from 34.5 (95% CI 33.4–35.6) to 32.3 (31.2–33.4) at 1y (p<0.0001)

Well-being scores increased (improved) from 26.8 (26.0–27.4) to 27.4 (26.7–28.1)(p=0.008).

Anxiety and well-being over 1y did not differ between participants receiving a normal or at-risk result
328 (76%) returned questionnaires at 1yNoneBMI and gender (more female) significantly different between groups, p <0.001 and p=0.002 respectively.

Same population as Farmer, 2005
Scientific Foundation Board of the Royal College of General Practitioners, funded by National Health Service Career Development Award
Farmer et al, 2005184
Fair
Randomised controlled trialTo assess the impact on response rates and psychological measures of different follow-up schedules in at-risk participants undergoing screening for DM2United Kingdom, Oxfordshire and South Northamptonshire431 total
Limited follow-up (LF): 213
Intensive follow-up (IF): 218
1m
6m
1y
Probands:
Aged ≥ 35 at diagnosis
Families with ≥ 3 siblings, and a quarter of families with 2 siblings living within study area
Participants:
Participants aged 35–74
Family history of DM2
Not known to have DM2
Able to complete questionnaires
Participants:
Known DM2
< age 35 or > age 74
Recruited with information from general practitioners
Probands sent questionnaires to assess willingness of siblings to participate
Mean age (SD):
LF: 58.8y (9.5y) IF: 58.1y (9.9y)
% Male:
LF: 48.8 IF: 42.7
Occupational group (manual / professional %)
LF: 61.4/81
IF: 63/37
YesNRPlasma glucose: (LF then IF)
Normal (<101 mg/dl [<5.6 mmol/L]: 112, 115
Borderline (101–108 mg/dl [5.6–6.0 mmol/L]) 50, 51
At risk (>108–<142 mg/dl [> 6.0–<7.9]): 43, 42
Diabetes (≥142 mg/dl [≥7.9 mmol/l]): 8, 10
NRNRBMI (mean):
LF: 27.7
IF: 28.6
Response rates calculated
Speilberger State Anxiety Inventory (SSAI-SF)
Well-being questionnaire (WBQ-12)
Random assignment to either limited follow-up (1y) or intensive follow-up (1m, 6m, 1y)

Analysis separated according to follow-up rates only
Response rates
Anxiety
Well-being
No significant difference between groups in SSAI-SF (anxiety) change scores from baseline to 1y follow-up (p=0.13)
Limited follow-up group had greater improvement in well-being (change score of the WBQ-12 well-being, p= 0.003
10% failed to return SSAI-SF follow-up
11.2% failed to return WBQ-12 follow-up
No difference between groups in proportion of 1y response questionnaires returnedIf group slightly more likely to be female, heavier, higher baseline WBQ-12 score

Focused on differences between 1 vs. 3 follow-up questionnaires, so groups not very meaningful for our purposes

Same population as Farmer, 2003
Scientific Foundation Board of the Royal College of General Practitioners, funded by National Health Service Career Development Award
Hoorn Study
Adriaanse et al, 2002181
Not rated
Cohort study with comparison (nondiabetic) group

(pilot study)
To explore psychological impact of a stepwise population-screening project for DM2Netherlands, Hoorn region40 total (11,679)
Diagnosed with DM2: 20
At increased risk: 20
Screen-diagnosed diabetes group: 2m
Elevated risk group (controls): 2w
Participant in Hoorn screening project and chosen to be part of pilot study
DM2 or elevated risk of DM2 (SRQ score > 6)
Ages 51–74
NRFrom population-based screening project; identified as high riskMean age:
DM2: 62.3y ± 5.9
nonDM2: 64.9y ± 6.2
% Male:
DM2: 50
nonDM2: 50

nonDM2 group was high risk
NRNR55% reported family history of diabetes in each groupFPG (mmol/l)
newly-diagnosed: 8.5 (2.3)
Non-diabetic: 6.5(0.6)
NRNRNonDM2 (N=20): 17 with IFG and 10 with both IFG and IGT

BMI:
DM2: 28.6 ± 3.5
nonDM2: 27.7 ± 4.1
SRQ - used to identify people in general population at increased risk for DM2

Semistructured interviews examining:
In newly-diagnosed DM2: the impact of diabetes, understanding of the test result, perceived severity, sense of control
In screened non-diabetics: impact of the test results, intention to change lifestyle
Both groups: views on the screening procedure
Qualitative study
Semi-structured interviews specific to intervention or control groups:
Newly-diagnosed diabetes group: 30–60 minutes at their home
Non-diabetic group: 15–30 minutes via telephone
Psychological impactScreening procedure: both DM2 and nonDM2 participants evaluated screening procedure as positive and not burdensome
1 person alarmed by diagnosis, the 19 others were not
Having diabetes was not experienced as a severe disease, no concerns were expressed
0Listed, but not standardizedWhen capillary glucose > 99 mg/dl (>5.5 mmol/L), venous FPG was measured and within 2w, a 75-g OGTT performed
Used WHO (1998) criteria (requiring FPG ≥126 mg/dl (≥ 7.0 mmol/L) on 2 separate occasions, or abnormal OGTT, with 2-h plasma glucose ≥200 mg/dl (≥ 11.1 mmol/L)
Health and Research Development Council of The Netherlands
Hoorn Study
Adriaanse et al, 2004178
Fair
Cohort study with comparison group (both with DM2)To determine prospectively health-related quality of life during 1st y following diagnosis of DM2, in newly diagnosed patients in general practice, compared with patients detected early by targeted population screeningNetherlands, Hoorn region165 total
GPDM (general practice diagnosed diabetes): 49
SDM (screening diagnosed diabetes): 116
2w
6m
1y
SDM: Participant in Hoorn screening project and chosen to be part of this study, with DM2, ages 50–75

GPDM: cities of Den Helder and Medemblik, 36 general practices, 1999-2001, with DM2, ages 50–75
NRFrom population-based screening project; identified as DM2

From general practices; identified as DM2
Mean age:
GPDM: 62.2±7.0
SDM: 63.2±7.3
% Male:
GPDM: 49
SDM: 56.9
Educational level:
GPDM: 57.1% low, 36.7% middle, 6.1% high
SDM: 62.1% low, 30.2% middle, 7.8% high

P value = 0.695, ns
YesSee “other results” column

Microalbuminuria (%):
GPDM: 26.5
SDM: 20.7

Impaired foot sensitivity (%):
GPDM: 51.0
SDM: 46.6

Retinopathy (%):
GPDM: 2.0
SDM: 8.6

Lipid lowering med (%):
GPDM: 16.3
SDM: 17.2
See “other results” columnNRNRSee “other results” column

BMI:
GPDM: 29.5±6.1
SDM: 29.7±4.9
SRQ - used to identify people in general population at increased risk for DM2.

Type 2 Diabetes Symptom Checklist (DSC-type 2) - measures presence and burden of diabetes-related symptoms

Short Form 36 (SF-36) - measures perceived health status

Well-Being Questionnaire (WBQ12) - Dutch version, measures emotional well-being
Completed standardized questionnaires at 2w, 6m, and 1y following DM2 positive test resultHRQoL, including: presence and burden of diabetes-related symptoms, perceived health status, emotional well-beingDSC-type 2 score (higher scores indicate more symptom distress) improved significantly within GPDM across follow-up (2w: 0.56; 6m: 0.21; 1y: 0.26, p<0.001), but not for SDM group (2w: 0.24; 6m: 0.24; 1y: 0.29, p=0.093)

GPDM consistently worse mean scores on all SF-36 mental health subscales and all WBQ12 scores at each time point compared with SDM Differences were statistically significant (worse) for GPDM group on SF-36 for Role Emotional (F=5.2, p=0.024), Mental Health (F=5.0, p=0.027), and Vitality (F=3.9,p=0.049); Significantly lower Mental Health Component Score for GPDM (F=7.0, p=0.009); Differences were statistically significant (worse) for GPDM group on WBQ12 for General well-being (p=0.048)

No differences between groups over time for other dimensions of SF-36 and WB12

SF-36 General Health (F=3.7, p=0.028) and Vitality (F=4.5, p=0.012) scores of GPDM improved significantly over time compared with SDM
GPDM: started with 71, data for 49
SDM: started with 217, data for 116
General practitioners reported that 76% (31/41) of newly diagnosed GPDM group were detected because of distinct diabetes-related symptoms

Baseline significant differences:
GPDM higher:
fasting plasma glucose (mmol/L) 9.7±3.1 vs. 8.5±2.0, p=0.005
A1c (%) 9.1±2.3 vs. 6.7±1.4, p<0.001
Oral blood glucose lowering agents (%) 77.6 vs. 24.1, p<0.001
SDM higher:
Overweight (BMI ≥ 25)(%) 72.9 vs. 88.8, p=0.011
Hypertension (%) 59.2 vs. 75.0, p=0.042
WHO (1998) criteria used for diagnosis

First study to compare these 2 groups
NR
Hoorn Study
Adriaanse et al, 2004180
Fair
Cohort with comparison (nondiabetic) groupTo examine impact of diagnosis of DM2 on psychological well-being and perceived health status in subjects who participated in a targeted population-screening programNetherlands, Hoorn region259 total (from 11,679)
Subsequently diagnosed with DM2: 116
Without DM2 143
2w
6m
1y
Participant in Hoorn screening project and chosen to be part of this study, with DM2 or elevated risk of DM2 (SRQ score > 6)
Ages 51–74
NRFrom population-based screening project; identified as high riskRace: >99% Caucasian
Mean age:
DM2: 63.2 ± 7.3
nonDM2: 62.2 ± 7.3
% Male:
DM2: 56.9
nonDM2: 51

nonDM group was high risk
NRYesParent or sibling with
DM2: 43.1%
nonDM2: 37.8%
FPG mmol/L
Diabetic: 7.3 (1.9)
Non-diabetic: 5.9 (0.3)
NRNRSignificant differences in BMI between groups: DM2: 29 + 5.1 vs nonDM2: 27.9 + 4.0, (p=0.045)SRQ - used to identify people in general population at increased risk for DM2

12-item Well-being Questionnaire (WBQ12) - Dutch version

Medical Outcomes Study Short Form 36 (SF-36)
Completed standardized questionnaires at 2w, 6m, and 1y following test result (DM2 diagnosis or not)Psychological well-being
Perceived health status
2w after diagnosis: no significant mean differences in psychological well-being nor perceived health status
6m after diagnosis: significantly lower scores of DM2 group for Role Physical (mean difference -8.2 [95% CI -16.2; -0.1], p=0.046) and Role Emotional (mean difference -7.9 [95% CI -15.3; -0.5], p=0.038) dimensions of perceived health status; no other significant differences
1y after diagnosis: no significant mean differences in psychological well-being nor perceived health status
NRNoneSignificant differences in BMI: DM2 29 ± 5.1 vs. nonDM2 27.9 ± 4.0, (p=0.045)

Use of antihypertensive drugs: DM2 36.2% vs. nonDM2 35.7%, NS.
NR
Hoorn Study
Adriaasne et al, 2005179
Fair
Cohort with comparison (nondiabetic) groupTo determine level of diabetes-related symptom distress and its association with negative mood in population-based screening program, comparing DM2 vs nonDM2 (but high risk) groupsNetherlands, Hoorn region246
DM2: 116
nonDM2: 130
2w
6m
1y
Participant in Hoorn screening project and chosen to be part of this study
With DM2 or elevated risk of DM2 (SRQ score > 6)
Ages 50–75
NRFrom population-based screening project, identified as high risk or DM2Mean age:
DM2: 63.2y ± 7.3y
nonDM2: 61.9y ± 7.3y
% Male:
DM2: 56.9
nonDM2: 50.8
Race: >99% Caucasian
NRYesNRFPG (mmol/l)
Diabetic: 7.3 (1.9)
Non-diabetic: 5.9 (0.3)
NRNRBMI (kg/m2):
DM2: 29.0±5.1
nonDM2: 28.0±4.0
SRQ - used to identify people in general population at increased risk for DM2

Diabetes Type 2 Symptom Checklist (DSC-type 2)

Negative Well-being (NWB) Subscale of Well-being questionnaire (WBQ12) - Dutch version
Completed standardized questionnaires at 2w, 6m, and 1y following DM2 Screening test
Analyzed all variables
Diabetes-related symptom distress
Negative mood
Screening-detected DM2 patients reported significantly greater burden of hyperglycemic (F = 6.0, p=0.015) and of fatigue (F = 5.3, p=0.023) symptoms in the 1st y following diagnosis; outcomes did not change over time, no significant group by time interactions were found

Total symptom distress (range 0–4) relatively low for both DM2 (median at 2w, 6m, and 1y; 0.24, 0.24, 0.29) and nonDM2 (0.15, 0.15, 0.18) and not significantly different

No average difference and change over time in negative well-being

Negative well-being significantly positively related with the total symptom distress score (regression coefficient beta = 2.86, 95% CI 2.15–3.58)
DM2: started with 156; data for 116 (74%)
nonDM2: started with 163; data for 130 (80%)
NoneNR
Nichols et al, 2004185
Poor
Cohort with comparison (nondiabetic) groupTo examine functional health status prior to diagnosis of DM2, and measure effect on functional health status of receiving the diagnosisUnited States
Kaiser Permanente Northwest, Portland, Oregon
Those meeting new diagnostic criteria (I): 498
Comparison group (C): 589

Originally 1014 in each group, response rate of 69%, missing items lead to final numbers (44%) N=273
1yMembers of HMO Kaiser Permanente Northwest
In Kaiser records, but not in diabetes registry, that meet new criteria for diabetes since ADA lowered diagnosis criteria from 140 to 128 mg/dl (7.8 to 7.0 mmol/l) (soon to be diagnosed)
Age and gender match comparison group without DM2
Previously diagnosed DM2Electronic registry database

DM2 vs nonDM2
Mean age: 66.9y + 10.5y
% Male: 56
NRYesSelf report:
Hypertension (p<0.001)
I: 61.6% C:38.7%
Heart problems (p<0.001)
I: 40.5% C: 23.5%
Neuropathy symptoms (p=0.003)
I: 30.7% C: 22.5%
Diabetes symptoms
I: 55.1% C: 47.8%
NRNRNRSelf report:
Depression
I: 14.1%
C: 13.4%
BMI (p<0.001)
I: 30.3%
C: 27.9%
SF-12 Health Survey
Physical component (PCS-12)
Mental component (MCS-12)
After ADA reduced fasting glucose level for diagnosing diabetes from 140 to 126 mg/dl (7.8 to 7.0 mmol/l) in 1998, searched Kaiser Permanente Northwest database back to 1994 (database started in 1988) identifying members who were not currently in diabetes registry, but that met new criteria (before diagnosis group) and added an age and gender-matched comparison group
Measured functional health status 1y before and 1y after diagnosis of DM2
Functional statusBetween-group at baseline:
Prior to diagnosis, physical functioning already lower in subjects who met the new criteria than comparisons (39.5 vs. 42.1, p<0.001); Mental functioning was ns (51.4 vs. 51.9, p=0.406)

Within-group after 1y:
Among those who newly met diagnostic criteria, no difference in change in health status (mental or physical) in those who reported receiving a diagnosis (n=105) compared with those who did not (n=168). Adjusted for age difference (at 1y follow-up) between those receiving diagnosis (younger) and those not (67.0 vs. 69.6, p=0.031);
After adjustment for age, learning of diagnosis was not associated with any difference in functional status on either questionnaire or with a change in physical (1.55 vs. 0.05, p=0.233) or mental (-0.63 vs. 0.01, p=0.598) health status compared to those who had not been told of their diagnosis
1y later: Sent out 706 follow-up questionnaires, 623 were still members, received 273 (44%) usable responsesThose meeting new criteria were more likely to report:
Hypertension (61.6 vs. 38.7%, p<0.001)
Heart problems (40.5 vs. 23.5%, p<0.001)
Neuropathy symptoms (30.7 vs. 22.5, p=0.003)
Diabetes symptoms (55.1 vs. 47.8%, p<0.019)
Higher BMI (30.3 vs. 27.9, p<0.001)
Adjusted for age difference at 1y follow-upNR
Peel et al, 2004186
Not rated
Cross-sectionalTo assess impact of DM2 new diagnosis on emotions and viewsUnited Kingdom, Scotland
Multicenter (16 different practices and 3 hospitals)
40No follow-upNewly diagnosed from range of backgrounds (poor, affluent, rural, urban) from various practices and hospitals across Lothian region in Scotland
Based within Local Health Care Co-operatives
NRRecruitment from general practitioners and hospitalsAge (mean [range]):
48y (21–77y)
52.5% male
47.5% female
Number of participants (using Registrar General's classification system):
Social classes I–II: 10
Social classes III non-manual: 12
Social class III manual: 13
Social classes IV–V: 5
NRPerhaps, but quantitative data NRNRNRNRPerhaps, but quantitative data NRIn depth interview (not standardized)In depth interviewEmotional reaction about diagnosis
Views about information provision at time of diagnosis
Varied emotional reactions to diagnosis
Most wanted detailed information at time of diagnosis
NANoneIdentified 3 “routes” to diagnosis:
1)

Suspected diabetes route

2)

Illness route

3)

Routine screening route

Scottish Executive Health Department
Skinner et al, 2005187
Not rated
Cross-sectional (1 time assessment at screening)To assess impact of diabetes screening on anxiety levels in ethnically mixed populationUnited Kingdom, Leicestershire1,339
1,189 (complete data sets)
No follow-upParticipant in Screening those at Risk (STAR) study
Ages 25–75 (40–75 if White) with ≥ 1 risk factor:
Known CHD, known risk of CHD or on CHD register, documented history of hypertension with medication, cerebrovascular disease and/or peripheral vascular disease, diagnosis of IGT or IFG, women with polycystic ovary syndrome and obesity (BMI > 25 or > 23 kg/m2 in South Asians, BMI > 30 kg/m2, BMI > 25 kg/m2 with sedentary lifestyle), women with previous history of gestational disease, first-degree relative with DM2
Housebound
Terminal illness
Previously diagnosed DM2
Unable to read or complete questionnaire unaided
Identified at high risk of developing DM2 though general practitioner's or cardiovascular team's lists, Coronary Heart Disease register, or through public media recruitmentHigh risk for DM2

54% male
46% female

21% Asian
75% Caucasian
4% Other

Ages:
Asian: 51.2y ± 11.2y
Caucasian: 60.5y ± 9.9y
NRNRNRNRTC Asian: 197±35mg/l (5.1±0.9 mmol/l)
HDL Asian: 46±15mg/l (1.2±0.4 mmol/l)

TC Caucasian: 209±46 mg/dl (5.4±1.2 mmol/l)
HDL Caucasian: 54±19 mg/dl (1.4±0.5 mmol/l)
Asian: 128 ±21/80±11 mmHg

Caucasian: 134±25/80±11 mmHg
Relative with diabetes:
Asian: 70%
Caucasian: 37%

BMI:
Asian: 26.88±4.4 kg/m2
Caucasian: 28.5+5.6 kg/m2
OGTT to assess diabetes status

To access anxiety: SSAI-SF, Emotional Stability Scale of the Big Five Inventory 44, and 3 scales from the Diabetes Illness Representations Questionnaire (modified for interviews)
Anxiety measured at time of screeningAnxietyNo effect of family history of diabetes ethnic group, or recruitment methods on anxiety
45% of participants reported “little to moderate” amounts of anxiety (mean 35.5, SD 11.6)
Emotional stability was significantly (negatively) associated with anxiety (r=-0.45; n=930; p<0.001), with females describing themselves as less emotionally stable than males (t=4.49; df=577; p<0.001)
There were no other variables significantly associated with anxiety
NRParticipants with a first-degree relative with diabetes were more likely to agree that diabetes was hereditary (t=3.22, p<0.001)

South Asians were more likely than Caucasians to agree that diabetes is hereditary (t=3.59; p<0.001) and caused by poor medical care (t=4.11; p<0.001), and less likely to agree that it is a chronic condition (t=3.38; p<0.001)

64% of responders thought diabetes was caused by diet
61% of responders thought diabetes was caused by hereditary factors
12% of responders thought that diabetes was serious, shortens life, and causes complications

Other outcomes relate to perceived causes of diabetes, duration of diabetes, and impact on diabetes on life
Cannot locate original STAR study

Issue with analysis, lost 150 datasets: “Because of problems with recording the ID number on questionnaires, a number of questionnaires could not be linked to results of standardized health assessment. Therefore, where data are reported that combines data from health assessment and the questionnaire, # of participants in analysis is substantially reduced.”

Authors described ethnically mixed population as 75% Caucasian 21% Asian; 4% Other
NR

Abbreviations: ADA, American Diabetes Association; ADDITION Study, Anglo-Danish-Dutch Study of Intensive Treatment and Complication Prevention in Type 2 Diabetic Patients Identified by Screening in Primary Care; BG, blood glucose; BMI, body mass index; C, control group; CHD, coronary heart disease; CVD, cardiovascular disease; DM, diabetes; DM2, type 2 diabetes mellitus; DSC-Type 2, Diabetes Symptom Checklist - Type 2 diabetes; FBG, fasting blood glucose; FPG, fasting plasma glucose; GPDM, general practice-diagnosed diabetes; HADS, Hospital Anxiety and Depression Scale; HAI, Health Anxiety Inventory; HDL, high density lipoprotein; HMO, Health Maintenance Organization; HRQoL, Health Related Quality of Life questionnaire; I, intervention group; IF, intensive follow-up group; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; LF, limited follow-up group; m, months; MCS, Mental Component Score; NA, not applicable; nonDM, without diabetes; NR, not reported; NS, not significant; NSD, no significant difference; NWB, negative well-being subscale; OGTT, oral glucose tolerance test; PCS, Physical Component Score; SD, standard deviation; SDM, screening-detected diabetes; SES, socioeconomic status; SF, short form; SRQ, Symptom Risk Questionnaire; SSAI-SF, Spielburger State-Trait Anxiety Inventory-Short Form; STAR, Screening those at Risk; TC, total cholesterol; w, week; WBQ-12, Well-being Questionnaire-12; WHO, World Health Organization; y, years.

From: Appendix B Evidence Tables

Cover of Screening for Type 2 Diabetes Mellitus: Update of 2003 Systematic Evidence Review for the U.S. Preventive Services Task Force
Screening for Type 2 Diabetes Mellitus: Update of 2003 Systematic Evidence Review for the U.S. Preventive Services Task Force [Internet].
Evidence Syntheses, No. 61.
Norris SL, Kansagara D, Bougatsos C, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.